INFORMATION TO USERS This reproduction was made from a copy of a m anuscript sent to us for publication and m icrofilm ing. W hile the m ost advanced technology has been used to pho­ tograph and reproduce this m anuscript, the quality of the reproduction is heavily dependent upon the quality of the m aterial subm itted. Pages in any m anuscript may have indistinct p rin t. In all cases the best available copy has been filmed. The following explanation of techniques is provided to help clarify notations w hich may appear on this reproduction. 1. Manuscripts may not always be complete. When it is not possible to obtain m issing pages, a note appears to indicate this. 2. When copyrighted m aterials are removed from the m anuscript, a note ap­ pears to indicate this. 3. Oversize m aterials (maps, drawings, and charts) are photographed by sec­ tioning the original, beginning at the upper left hand com er and continu­ ing from left to rig h t in equal sections w ith small overlaps. Each oversize page is also film ed as one exposure and is available, for an a d d itio n a l charge, as a standard 35m m slide or in black and w hite paper form at, * 4. Most photographs reproduce acceptably on positive m icrofilm or m icro­ fiche b u t lack clarity on xerographic copies made from the m icrofilm . For an additional charge, all photographs are available in black and w h ite standard 35m m slide form at. * ♦For more information about black and white slides or enlarged paper reproductions, please contact the Dissertations Customer Services Department. T3mversity M crafibiis International 8603445 Lindquist, Kirk Lars BUDGET REDUCTION DECISIONS IN STATE GOVERNMENT: STUDY OF MICHIGAN, 1979-1983 Michigan State University University Microfilms International 300 N. Zeeb Road, Ann Arbor, Ml 48106 Ph.D. A CASE 1985 PLEASE NOTE: In all cases this material has been filmed in the best possible way from the available copy. Problems encountered with this document have been identified here with a check mark V 1. Glossy photographs or pages_____ 2. Colored illustrations, paper or print______ 3. Photographs with dark background_____ 4. Illustrations are poor copy______ 5. Pages with black marks, not original copy______ 6. Print shows through as there is text on both sides of p a g e _______ 7. Indistinct, broken or small print on several pages 8. Print exceeds margin requirements______ 9. Tightly bound copy with print lost in spine_______ . J 10. Computer printout pages with indistinct print______ 11. Page(s)____________lacking when material received, and not available from school or author. 12. Page(s)____________seem to be missing in numbering only as text follows. 13. Two pages numbered 14. Curling and wrinkled pages______ 15. Dissertation contains pages with print at a slant, filmed as received_________ 16. Other___________________________________ _________ . Text follows. ______ University Microfilms International BUDGET REDUCTION DECISIONS IN STATE GOVERNMENT A CASE STUDY OF MICHIGAN 1979-1983 By Kirk Lars Lindquist A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY School of Urban Planning and Landscape Architecture 1985 ABSTRACT BUDGET REDUCTION DECISIONS IN STATE GOVERNMENT A CASE STUDY OF MICHIGAN 1979-1983 By Kirk Lars Lindquist Few studies of government budget crises have been done which might help explain how reduction decisions are made. If the effects of actual program cuts are not those expected by decision makers, it is difficult to determine the reason for the difference. A model is needed which can help explain the process of budget cutting. Before such a model can be proposed, it is first necessary to describe how these decisions are made. The purpose of this research is to establish a framework for analyzing reduction practices and to describe the budget reduction process. This case study describes and analyzes the process followed by the State of Michigan as it attempted to resolve its revenue shortfall problems during the budget years between 1979 and 1983. It also reviews the types of information requested by budget officials. An attempt has also been made to assess the relative influence of various budget officials. Information used in this case study of Michigan was gathered through personal interviews and a short survey. Each included open Kirk Lars Lindquist ended questions designed to shed light on the way that decisions were reached between 1979 and 1983. This study found that the reduction process used in Michigan did not correspond to any single pattern proposed by process, comprehensive, or incremental theorists. rational/ As reduction plans were prepared, some agencies used program priorities and performance criteria to make their decisions while others distributed reduction targets on a fair share basis. Different groups of officials dominated at various points in the process. Agencies, the Department of Management and Budget, and the Legislature each directly influenced the content and consequences of program service cuts. to be more influential than any other. No single person or group proved ACKNOWLEDGMENTS I am indebted to many people for their assistance and guidance. However, there are several individuals who played a crucial role in the preparation of this dissertation. I would like to thank my guidance committee: Honey, Lawrence Libby, and Milton Steinmueller. organization and content, and penetrating Roger Hamlin, Keith Their suggestions on questions have proved invaluable in developing this dissertation from concept to this final product. I am indebted to Nora Beckett for her able clerical assistance in the preparation of this final document and to Sharon Telleen who provided important editorial guidance. I would like to give a special thanks to my wife Barbara and my eager assistants Anders and Britt. Without their patience and constant support none of this would have been possible. i TABLE OF CONTENTS Page List of Tables.......................... iv List of Figures.................... v I. INTRODUCTION............................... ....... Overview.......................................... Problem........................................... Purpose of Research............................... Organization of Study............... ...... 1 1 2 4 7 II. ALTERNATIVE PERCEPTIONS OF THE BUDGETARY PROCESS......... Introduction............................. Budget Theory..................................... Rational/Comprehensive Model....................... Incremental Budget Models.......................... Process Models of Budget Behavior.................. Roles and Influence............. Budget Theory and Reduction Decision Making......... Possible Reduction Decision Patterns................ Summary........................................... 8 8 12 17 21 22 26 28 30 III. METHODOLOGY ................................... Introduction...................................... Hypothesis........ Michigan Case Study................... Research Design................................... Evaluation of Results.............................. Limitations of this Research Design................. 35 35 36 38 39 44 46 IV. MICHIGAN 1979-1983....................................Overview...... The Legacy of the Program Budget Evaluation System.... Spending Patterns................................. Institutional Responses to Fiscal Stress 48 48 57 61 63 V. HOW REDUCTION DECISIONS WERE REACHED.................... Introduction...................................... Executive Order Process............................ DMB Phase......................................... Effects of Structure on Strategy....... Program Impact Information.................... Legislative Phase................................. 73 73 74 77 86 89 90 ii Constitutional Provisions.......................... VI. SUMMARY AND CONCLUSIONS................................ Introduction...................................... Interview Results ..... Strategy..................................... Decision Units................................ Issues Development................. Budget Official Influence..................... Dominance........... Assignment................................... Political Variables................................ Impact Information............................ Agency Prosperity............................. Distribution of Cuts ................... Institutional Variables............................ Agency Size......... Program Complexity ..................... Budget Structure .................... . Economic Variables ........................ Propositions ................................... Summary........................................... 93 98 98 99 102 102 103 104 104 105 107 107 108 108 109 109 109 109 110 110 116 APPENDICES.................................................. A. State of Michigan Executive Order Summary............. B. Reduction Procedures................................ C. Letter of Introduction and Questionnaires....... D. Executive Order Summary Selected Departments.......... 118 118 120 122 130 SELECTED BIBLIOGRAPHY....................................... 131 iii LIST OF TABLES Page Alternative Operational Measures................... Alternative Budget Reduction Models................. Evaluation Framework.............................. . Interview Results Worksheet........................ Budget Process Participants........................ Budget Process Roles.............................. Categorical Spending Patterns 1968-1978............. Michigan vs O.S. Unemployment Rates 1956 to 1984.... Wage and Salary Employment in Michigan Selected Years 29 41 42 45 49 49 62 1 9 6 0 -1 9 8 2 ..................................... 67 68 General Fund-General Purpose Year End Fund Balances Department of Natural Resources Forestry Appropriation Unit Detail....................................... Interview Results. ............................. Michigan's Reduction Process....................... iv 66 83 100 101 LIST OF FIGURES Page 2.1 4.1 5.1 5.2 6.1 6.2 6.3 Framework for Budgetary Analysis........................ Role Premises: Michigan’s Executive Budget Process....... State of Michigan General Fund Balances................. Michigan's Budget Reduction Process: Department of Management and Budget Phase............................. Relationships and Outcomes of the Executive Order Process.. Group Influence C/er Reduction Order Outcomes............ Influence and Reduction Order Content................... v 11 52 78 79 111 112 115 CHAPTER I INTRODUCTION Overview State and local governments operate within a set of finite fiscal limits and revenue levels define these limits. Most jurisdictions adopt budgets or spending plans to assure that their priority programs and services will be provided. same time. Not all demands or needs can be met at the A budget reflects an agreement between a government and its constituency. Budgets therefore are a political contract and a legisla­ tive mandate to use tax and other revenue within specific expenditure limits for specific public purposes. Two general assumptions are evident in budget decisions. One suggests that it is the responsibility of legislative bodies to provide enough sufficient levels of revenue to allow government to respond to citizen needs. This can be difficult to do when the tax base is no longer growing. The other assumes that the size of government should be restrained by the amount of money it receives. In both cases, it is the task for budget officials to be sure that priority needs are supported. State budgets are complex documents. Some revenue sources are tied to specific services and programs while others may be used for general purposes. These non-restricted revenue sources often support 2 most budgeted items. Often referred to as General Funds, these revenue sources are especially sensitive to changes in local and state economic conditions. Governments with constitutional provisions for balanced budgets must make adjustments as shortages occur to keep government spending in line with expected revenue levels. Most state governments have been confronted with the difficult political task of reducing appropriated funds. adjustments are relatively minor. In most cases, these Shortages are usually in the neighborhood of two to three percent of total appropriations and are handled administratively by the state budget officer. Provisions for legislative oversight vary from state to state, but revenue reconcili­ ation is generally an executive responsibility. Legislative involvement is more evident when these shortages become more pronounced especially when reductions of ten to fifteen percent are needed. The principal issues addressed during the budget process relate to the size of programs and to the distribution of services. Analysts try to discover the minimum operating levels for programs out of a general concern that basic services will continue to be provided. Both budget allocation and reduction decisions focus on program needs and demands for services, but we do not know how these are taken into account. The budget reduction decision making process is not sufficiently understood to enable theorists to model the process. Problem Most studies of budget decision making have analyzed the federal budget process. Some researchers have analyzed budget practices, but these studies either describe decision making patterns or attempt to predict appropriation levels. protracted budget published. No organized study of a prolonged and crisis in state or local governments has been Consequently, there is no basis for suggesting what a normal response to a twenty percent revenue shortfall might be. It is important for government officials to understand how budget reduction decisions are made. Such an understanding should also be helpful to those who study and analyze the practice of budgeting from an academic perspective. Just as Suchman suggests, if one is to appreciate how the budget process works, stress. it can be helpful to observe it while it is under Stressful conditions will force "weak" points of the process to bend and adjust to these forces. It can be very instructive to see how a government changes its established procedures in response to problems not encountered before. One can observe the maturation of budget decision practice. Three different descriptions of the budget process have been proposed: a rational/comprehensive model, a process model and the incrementalist model. Rational/comprehensive model advocates maintain that program priority and performance issues should dominate budget decisions. They also feel that the roles and responsibilities of officials and the sequence of these decisions are important independent variables. Process theorists agree with these role and sequence propositions and suggest that appropriations and spending plans reflect the preferences of these officials. sequencing is important, but suggest Incrementalists agree that the single most important determinant of an appropriation is the prior year funding level and the way that targets are assigned in the political arena. No study has sought to reconcile these perceptions of the budget process. No theoretical base appears to be available to assist in predicting how state governments reduce their budgets. Before any description of reduction decision making patterns is possible, the appropriateness of these three models should be explored. Through an organized and thorough analysis of reduction practices, propositions can be offered that can suggest how reduction decisions are made. Purpose of Research The objective of this study is to describe the budget cutting process, show how the process works, and illustrate how the State of Michigan reached its financial goals during a recent period of major revenue decline. The processes and techniques used by budget officials to decide which programs will be reduced, or eliminated completely are analyzed. These practices are then compared to accepted models of budget decision making. The Information and conclusions of this study could be useful to budget officials contemplating massive program cuts and to academics concerned with the behavioral aspects of budget decision making. It is also a point of reference for other studies of reduction decision processes. This case study seeks to provide answers to several questions which might help explain the process of budget reduction. These relate to where reduction decisions were made, who made these decisions, how these reductions were made, and why these decisions were made as they were. Where Were Decisions Made? Are most budget decisions made in the agencies while only incremental changes are made by the Governor and the Legislature? Are reduction packages reviewed in detail by the Governor (via DMB) and the Legislature or does a different decision making pattern exist? different kinds of decisions made in the agencies, legislature. Are DMB and the If so, what influence might this pattern have on final reduction decisions? Bho_Makes Reduction Decisions? It is fairly easy to list or identify those who make reduction decisions. Problems come when one attempts to explain these agreements in the context of the roles and relationships among these individuals at various stages of the process. Are the relationships between department directors and the Governor more important than the influence of the appropriations committees in explaining reduction outcomes? Do these relationships change as successive stages of the process are reached? How are roles and relationships influenced by the bureaucratic and political environments of the budget process? How do changes in social, political and economic trends influence decision levels, roles, planning horizon and the types of decisions made? How Are Reduction Decisions Made? Are budget cut decisions made in the same way as those which resulted in non-marginal or non-incremental program increases? Are these decisions based on problems, analyses of alternative ways of resolving these problems and administrative efficiency? If so, why; if they are not, why not? Is it true that reductions are made through a rational process of choices? Should fiscal managers be expected to rank objectives and 6 resources together as reduction decisions are made? When minor revenue adjustments are needed, are decisions made at lower levels (positions, of the department organization focused on inputs travel expenses and infrastructure needs)? Are major cutbacks decided as packages of smaller decisions on a program basis? Do these decisions lead to programmatic choices through a process that prioritizes government activities? Why Are Reductions Made as They Are? Rational/comprehensive behavior might be a function of the duration and severity of the fiscal crisis. If long term cuts are needed, do budget analysts ask for and use program impact and performance infor­ mation? If modest cuts are needed, is input or performance information more important to budget officials? It is also proposed that most state government budget decisions are made in the agencies while only marginal or incremental changes are made by the Governor and the legislator. If process theorists are correct, the roles and relationships of budget officials are important variables in the reduction process. Reduction decisions may be a result of a budgeting-by-priority process and reflect balance of influence within government. It has been suggested that some of the techniques used to prepare reduction plans are undesirable, yet they have been used. An improved understanding of the relationships and roles of participants in the process and the effect review procedures have on budget decisions is important if a model of the reduction process is to be developed. Organization of Study No state government has experienced a budgetary crisis as severe and for such a long period as Michigan did between 1979 and 1983. dissertation documents how these reduction decisions were made. This Chapter II presents the theoretical basis for this study of Michigan’s fiscal crisis. Through this literature survey, three models of budgeting are presented as alternative explanations of budgetary outcomes. These are presented as different ways of looking at the reduction process. This chapter also offers three different theoretical propositions for how budget reduction issues might be presented and resolved. Chapter IV describes the budget process and relationships of different officials and interests. the roles and This chapter also offers an historical explanation for patterns of behavior evident between 1979 and 1 983. A number of innovative approaches to budgeting were attempted during the early and middle 1970's, and these experiences influenced how reduction issues were resolved. Chapter V describes how reduction decisions were made in Michigan through the eyes of various participants. It explains how these officials interacted and what decisions tended to be made at various points in the executive order process. The final chapter describes patterns which emerged and offers a model of the reduction process. CHAPTER II ALTERNATIVE PERCEPTIONS OF THE BUDGETARY PROCESS Introduction The rational/comprehensive, incremental and process models each offer explanations of budget reduction behavior and outcomes. Whether one model explains the reduction process better than another is influenced by the focus of the model. Do those involved in a decision focus on aggregates or on the detailed items of expenditure? Lance LeLoup suggests that budget decision making is a result of deliberate changes made in the budget process. With centralization of budget decision making comes a greater emphasis on aggregates. In comments made at a workshop conducted as a part of the American Society for Public Administration 1984 National Convention in Denver, CO., Dr. LeLoup suggested that there might be evidence at the state and local government level of a shift from micro to macro budgeting. This "shift to the macro" has brought several important changes to the practice of federal budget making. Congress has become more fragmented and weaknesses in the system are becoming more apparent. opportunities This has increased for some interest groups to influence the content of budget recommendations, while others have lost ground. 8 Information problems have developed. Baseline and assumption issues have fooused attention on interest rates and beneficiaries instead of program effectiveness (given levels of expenditure) and staffing patterns. Another member of the same panel, Louis Fisher [Congressional Research Service, Library of Congress] believes that many of these problems have developed as a consequence of the centralization of review of revenue and expenditure policy in the Congressional Budget Office, and the way that these issues are presented to the Congress as a whole. He suggests that no deliberation occurs on revenue or debt limit matters and contrary to intent, budget bills are passed and sent up for signature accompanied by a revenue/budget target resolution containing recommendations based on information more than one and one-half years old. "There are no incentives for rational decision making. There is no sense of responsibility and there is no interest in restraining spending or holding the system responsible. worlds." It is the worst of all (Colorado, 1984). LeLoup centralized. also suggests that program cut decisions tend to be This increases the likelihood that the fiscal detail will not shape final decisions. These decisions depend on established priorities and define revenue ceilings for planning purposes. These issues must be resolved before restraints on agency spending plans can be enforced. In practice, one might expect to find both program and policy decisions consequences depending on the organizational environment and the these decisions might have for the government as a whole. LeLoup's proposed framework for budgetary analysis is shown in Figure 10 2.1. He develops a classification of decisions categories — consisting of three priority, program, and operation decisions. Priority decisions shaped broad issues and are reflected in the functional allocation of the general fund budget among agency programs. Nearly all state budgets specifically address program needs. These "program decisions" cover specific areas within agencies and may include entitlement programs, construction projects, and expansion requests for existing programs. "Operations decisions" tend to be micro in focus and are evident in agency requests, expenditure patterns, and specifically deal with the financial requirements of departments. This priority- program-operation classification might be compared to the PlanningManagement-Control typology proposed by Robert Walker and popularized by Schick, Schutz and others. Three variables should be considered in the course of studying budget decision making: the key actors, key relationships and environ­ mental factors which might affect these decisions. Decisions tend to be dominated by a leadership group and are influenced by decisions already made. Higher level (macro) decisions establish constraints for lower (micro) level decisions. Micro-decisions tend to be summarized as they are presented to higher level governmental leaders. All decisions are influenced by non-institutional environmental factors. Economic trends can have a more significant impact than social or political in that these will influence the amount of revenue support available. A state's economic health and welfare is reflected in demands for government services caused by changes in inflation and unemployment rates. Figure 2.1 FRAMEW ORK FOR BUDG ETARY ANALYSIS * Constraints Decision Levels Priority Social Macro Most Prominent Actors Governor and Advisors Bureau o f the Budget Main Time Reference Annual Legislative Appropriation Committees Types of Decisions Statewide Spending Levels and Lapse Totals Rising Levels of Aggregation Economic Program M acro/M icro Agencies, Interest Groups Bureau o f the Budget Legislative Fiscal Agencies and Approp. Subcommittees t Annual and M ulti-year Open-ended Requests, Allocations, Capital Outlay Continual with a Midyear Review of Expenditures Amount, Timing, Transfers, and Control of Spending Political Operations Trends Micro Agencies Approp. Committees & Subcommittees Departm ent of Management Budget Perceived Severity of Cuts * Adapted from Lance T. Leloup, "The Myth o f Inerementalism: Analytical Choices in Budget Theory" 12 This framework offers a context for observing reduction behavior patterns. Explanations can also be offered through three models of budget decision making: Incremental, Rational/Comprehensive and Process. Budget Theory Before any principles of budgeting can be established, it is first necessary to describe how the process has been designed and to explain how it works.** It is fairly easy to list the major efforts at system description. The problem comes when one attempts to definitively describe the budgetary process in theoretical terms. Now, twenty years after the efforts of Wildavsky, Davis, Anton, Sharkansky and others, there is still disagreement over normative questions of efficiency and rationality. There is no agreement or the basic evaluative units — should they be outcome (i.e. dollars), or process variables (e.g. roles, relationships, structures and environmental influences on the task environment). Normative arguments decrying the irrationalities of politics have been neutralized at least momentarily with the demise of MBO, program, and zero-based budgeting. The fact that nothing has risen to take their place indicates that a certain amount of re-thinking may be occurring about what might constitute desirable budget decision practice. While budgeting is essentially administrative, decisions are reached through an unstructured political process. Three different approaches to the theory problem of budgeting are offered. Some describe the practice of budgeting in process terms, others feel that budget decisions should be the product of a rational/comprehensive analytical process, and proponents of an incremental view of decision making seek to predict future budget decisions based on past funding levels. All acknowledge the importance of political bargaining in any model of the budgetary process. They differ in focus, and consequently offer different perspectives of the way that budget decisions are made. The starting point for most budget theorists was provided by V.O. Key in 19^0. He proposed that writers had missed a basic point as they analyzed procedures and organizational needs, formats of budget requests and other "mechanical foundations for budgeting." He felt that the allocation of public funds was the most important issue. "On what basis shall it be decided to allocate X dollars to activity A instead of activity B"? Key argues that the answer can only be approached from the perspective of political philosophy: "The most advantageous utilization of public funds resolves itself into a matter of value preferences between ends" with no accepted formula which might 'determine* how allocation should be made.^ Some economists conclude that intrinsic and extrinsic value sets independently influence the distribution of resources. Scholars have sought to explain why budgetary officials act differently in seemingly identical decision making environments. In response to Key’s challenge, the three models of budget decision making have evolved. Each identifies variables which influence the behavior of participants in the budgetary process. Activities have been classified and linkages explored to better describe the structures and roles of incentives. two-fold: Their approach to the budget theory problem generally is what are the standard operating procedures, norms of conduct, and "unofficial rules of the game"? and answer to the question "what 14 behavior is evident as changes occur in the rules"? How do participants interact in budget decision making, influencing outcomes, and how might the behavior be influenced by political, economic and social factors. At the time Key issued his challenge very little was understood about the budget process. Fisher (1980) observes that public policy can be implemented only if the planning process successfully identifies and describes how environmental variables affect conditions for goal a t t a i n m e n t .3 To the extent that budget planners fail to observe the relationships of variables environment influencing or characterizing the task (in which a program is to operate), their plans for implementation may be invalidated or made irrelevant. Rational/Comprehensive Model The discipline of economics suggests that budget decisions should be the product of a rational process of choice among alternative means delivering public services based on norms of efficiency and effective­ ness. Once an understanding is reached of the consequences of each of several alternatives, a plan is selected which is the least expensive yet achieves basic program objectives. The rational/comprehensive model suggests that budget decisions should be based on: 1. a comprehensive overview of factors relevant to the decision. 2. a definition of social objectives in clear and useful terms. 3. a means-ends approach to policy. 4. a deliberate and explicit choice among policies 5. a calculation of minimizing costs. 6. a reasoned and cooperative dialogue opposed to arbitrariness coercion or conflict. 15 7. a unified decision-making process in those cases where these decisions are highly interdependent. This model assumes that voters and policy makers have relatively complete knowledge about what individuals and groups want. It also expects that objectives can be reached through a process of budget allocation that incorporates orderly analysis of alternative program means of service delivery. It assumes that voters and policy makers think and generally want to act rationally. Proponents feel that these "choice" decisions should be centralized in the group of select govern­ ment and political officials and not left to an irrational political process. Budget process innovations developed out of this rational comprehensive paradigm are designed to de-politicize the budget process - to make it more rational. The Executive Budget concept of budget making assured the preeminence of the executive branch in the preparation and justification of budget recommendations to the legislative branch. The President or the Governor provided policy guidance and to the agencies and congress (or legislature) for government as a whole. Before this innovation was made, budget allocation decisions were made through agency-legislative compromise and reflected the judgments of the appropriations chairman more than welfare concerns of the public-at-large. Other budget innovations were designed to enhance the analytical quality of budgets by institutionalizing principles of administrative efficiency. Herbert Simon (1945) proposed that budget decision making should be guided by these rational models. A fundamental principle of administration, which follows almost immediately from the rational character of "good administration," is that among 16 several alternatives involving the same expenditure the one should always be selected which leads to the greatest accomplishment of administrative objectives; and among several alternatives that lead to the same accomplishment the one should be selected which involves the least expenditure. Since this "principle of ef fi ci en cy" is characteristic of any activity that attempts rationally to maximize the attainment of certain ends with the use of scarce means, it is a characteristic of economic theory as it is of administrative theory. The "administrative man" takes his place alongside the classical "economic man. The efficiency principle was institutionalized in 1965 when the program budget concept was implemented throughout the Federal Govern­ ment.The norms of rationalism routines and documentation System (PPBS). were prominently displayed in the evident in the Plannlng-Programming-Budget In his testimony before the Subcommittee on Economy in Government (September 21, 1967)# Otto Davis expressed his preference for this innovative process. "I think that if one is concerned with obtaining a rational allocation of resources, or getting the most out of what we have, then one can only sit back and applaud the basic objectives of PPB...the careful definition of objectives, the development of plans for the future and the projection of budgets to get the future costs of programs, the identifi­ cation of programs and the careful stating of the budget in terms of program costs rather than by line item, the emphasis upon quantitative analysis, the explicit consideration of alternatives, the expanded role of cost-benefit analysis."6 Program budgeting has been described as a decision making structure and therefore it assumes that program decisions have been arrived at through this same process. Just as the Federal Government experimented with PPBS, Michigan implemented its own Program Budget Evaluation System (PBES) in 1973-71*. While it has also abandoned the program budget approach to budgeting, 17 many of the vestiges instructions, of the system are documentation, current budget process. still apparent in the and decision making frameworks of the This might be contrasted with the incremental and process models which suggest that budget outcomes are selected through a political process. The way that these budget decisions are made will determine what decisions will be reflected in agency spending programs. The systems approach to budgeting could not change budget decision making practices. While budget decisions are often deliberated in program terms in the executive branch, they are often reached in a more modest and incremental way by legislative bodies. Incremental Budget Models In 1959, Charles Lindblom presented a rational/comprehensive model and contrasted it with a model of decision making which recognized the limited abilities of most officials to meet the rigorous demands evident in the administrative efficiency literature. the product of successive limited Decisions, he argues, are comparisons. The rational/ comprehensive approach to problem solving is not feasible. be practiced except for relatively simple problems. "It cannot It assumes intellectual capacities and sources of information that men simply do not possess.Successive limited comparisons are more feasible and rational than comprehensive analysis. Using a tree analogy, it is not possible to start at the roots or to look at a situation from the ground up in determining causes or defining the contextual basis for initiating the program in the first place. It makes much more sense to start at the branches...to start with the current situation and selecting that 18 information which succinctly and clearly presents those policy options which can be exercised easily and will give the necessary relief. The decision maker may therefore select the first feasible choice he comes upon. Policy evolves out of a series of trials, errors and revised i trials, each modifying or building upon decisions already made. This helps avoid lasting mistakes which otherwise might have been made if a comprehensive and long range focus had been used. Advocates of the incremental method of policy formulation prefer it over the comprehensive approach both on procedural as well as on normative grounds. Lindblom suggests that it not only is the common way of doing business, it is the preferred way. Incremental steps can be taken quickly because they are only incremental. Further they do not tend to generate controversy as readily as other more dramatic changes. Political interaction enhances the supply of program information and the rationality of the process. It represents a method of conducting political business and therefore a strategy for bringing policy changes through a series of nudges and shoves without the pain and drama of explosive new initiatives. Several authors argue that incremental models best explain budget outcomes. A budget for a given year is marginally different from that of the year before. In terms of appropriations, budgeting is always an incremental process. The size of the increment or margin may change but little else. Traditionally, analysis has been directed to the marginal changes caused by the collective efforts of the agencies to expand their basic or continuing level programs. Wildavsky, Fenno, Sharkansky, Lindblom’s concepts of partisan Hrebrenar and others have related mutual adjustment and incremental 19 decision making to the manner that issues are deliberated and resolved during the budget process. They feel that the most significant determinant for agencies budget is the prior year's budget. Budgeting is incremental, not comprehensive. The beginning of wisdom about an agency budget is that it is almost never actively reviewed as a whole every year in the sense of reconsidering the value of all existing programs as compared to all possible alternatives. Instead, it is based on last year's budget with special attention given to a narrow range of increases or decreases. This conclusion is supported by the empiricalevidencegathered by Davis Dempster and Wildavsky and Richard Fenno in their studies of appropri­ ations politics and Congressional budget calculation practices, decisions and outcomes.^ Incrementalists feel that a budget is not the product of a rational/ comprehensive method of decision making, as Simon and other suggest it should be. Wildavsky feels that it is irrational to follow the comprehensive models evident in program budget practices. choice limits calculation so choices can be made, uses "Rational history to correct mistakes, harnesses power through organizational incentives and never, never ranks together."^® objectives or resources above but always Incrementalism facilitates decisions by simplifying and limiting conflict to those areas where major changes are requested. It facilitates the political process-logrolling and bargaining while deemphasizing overt conflict. In such a setting the political process guides decisions, not analysis. ["Rational" budget] systems are defective because they constitute bad advice on what is (and how to get) good policy analysis. Analysis is about action as well as advice; therefore, analysis, judging information good only if organizations actually use it to do better, seeks to embed itself in organiza­ tional incentives. Analysis is historically 20 grounded to use conflict creatively, for otherwise, by magnifying disagreement, it becomes its own worst problem. Analysis does not consider alternative objectives or alternative mixes of resources in isolation but rather together as alternative programs.11 The incremental concept has been applied to two different aspects of the budget process: the decision making process of mutual adjustment and bargaining and the final appropriations decisions themselves which result from that process. •Lance LeLoup observes that this concept has been further refined as a basis for predicting outcomes. Decision rules have been translated into regression equations (Davis, Dempster & Wildavsky) and work continues in refining the econometric models proposed by Davis, et. al. (1966), Wanat (197^), Padgett (1980) and Gist ( 1982). The incremental model that evolved out of this research fits congressional behavior much better that it does that of the executive and the agencies. Davis, in his testimony before the Subcommittee on Economy in Government, described the original model this way. "If one looks at the fits of the equations, one is struck by the fact that the fits for congressional equations are just huge orders of magnitude better than they are for the agency equations...Congress follows this creeping incrementalism much more than do the agencies and the Bureau of the Budget." Assuming for the moment that reduction decisions are arrived at through a rational/comprehensive process in the agencies and the legislature continues to use traditional methods of calculating agency appropriations, no added control over budget decisions and the reduction process will be realized by the legislature through a program budget system (testimony September 21, 1967). 21 Efforts are being made to build procedural aspects of the budget process into the equations to improve their predictive power. These efforts may prove to be inconclusive given the variable characteristics of budget process as one looks at different situations, governmental units and levels of budget analysis. Process Models of Budget Behavior Budget issues are addressed by a variety of officials in different administrative environments subject to influence by divergent factors arising from social, economic, political trends. Process theorists suggest that these incremental models do not sufficiently explain how budget decisions are made. A budget is the product of a process. A model designed to represent budget decisions as a consequence of procedural variables must deal with a number of highly complex environments. Roles and relation­ ships of various budget officials change as one proceeds from one phase in the cycle to the next. roles, Decisions become the units for analysis and constraints and organizational settings act as vehicles for shaping the level of that analysis. It is assumed that budget decisions can be explained if one knows (1) how an organization functions and (2) is aware of environmental inputs. Methodologically, the process theorist might first develop a set of rules which might explain how decisions are made and then to predict budget propositions. levels from these Gerwin’s study of sources of organizational stress and Anton’s study (1966) of the effects of changing revenue levels on agency spending decisions showed how procedural changes on budget decisions could be analyzed. 22 Natchez and Bupp (1973) found that budget decisions are the product of a budgetary process and a policy process. different meanings in each context. Appropriations take on They also believe that the program director is the central figure in the determination of budget policy. "Programs prosper," they say, "because energetic division directors successfully build political support to withstand continuous attacks upon a program's resource base by competing claims. As a consequence, only matters which reach the president are those already in dispute."^ Programs lacking this support are vulnerable and may be swept away in a budgetary crisis. Roles and Influence The standard definition of roles as most people use it refers to a pattern of behavior and usually in a dramatic sense. As Simon says in his introduction to the second edition of Administrative Behavior, "role" implies a specific pattern of behavior and is only meaningful when two or more people interact. Social role is adaptive to and contingent upon the situation and environments of these interactions. Simon defines role as a specification of some, but not all of the premises that enter into an individual’s decisions. "Behavior can be predicted, then, when the premises of decisions are known (or can be predicted) in sufficient detail." The decision premise is the appropriate unit for the study of administrative decision making he says and offers a basis for comparing different decisions.^ Situations are independent variables in studies of decision making since they select out decision premises. 23 The roles various budget officials assume at different phases of the budget process will influence system behavior and changes in a person’s role can influence final budget outcomes. An official may act as a budget reviewer when dealing with program requests of subordinate divisions, reviewing, discarding, modifying or adopting recommendations placed before him. However, this same official will act as a planner- prioritizer as a budget strategy is prepared for a consolidated request and as an advocate as a request is presented to the next level of authority in the organization, or at the next step in the process. The relative influence of budget officials is a function of their roles, the administrative structure and the environment at that point of time, place and given the agenda for discussion. Process theory adherents might agree with incrementalists on matters relating to the political aspects of the bargaining process. their explanations of these decisions. Where they part ways is in Incrementalists see the budget process as a series of bargaining exchanges in the agencies, among bureaus and departments, between departments and chief executive and before legislative appropriations subcommittees. These typically are between two individuals where one has hierarchical authority over the other. Outcomes are measured in terms of how well they do against the higher level in the hierarchy.^ While these exchanges are acknowledged by process advocates, they see much more occurring at the bureau and division level within these organizations. levels. These can be between equal In those conditions where one bureau's growth will result in anothers decay, a considerable amount of discussion may be expected (Natchez & Bupp, 1973). critical variables. Internal environments, roles and strategies are As much as ninety percent of the budget decisions 2H evident an approved budget might be made within the agencies. The final word on the allocation of appropriations among bureau programs might be expressed by department directors, division chiefs or even by the program managers themselves. Allen Schick suggests, "while the chief executive is usually portrayed as the person who makes the budget, agencies are collectively the most influential parties to a government's budget."^ The strategies used by departments or agencies in efforts to secure favorable budget allocations may take several forms. Incrementalists suggest that they would be well advised to ask for moderate increases. Trust might be lost if an agency predictably asks for large increases in its programs. In The Politics of Budgetary Process. Wildavsky suggests that agencies will be punished if this behavior goes on too long. If an agency continually submits requests far above what it actually gets, the Budget Bureau and the appropriations committee lose confidence in it and automatically cut large chunks before looking at the budget in detail.^ Other research suggests however that agency assertiveness may have little to lose and everything to gain by adopting an expansionary policy (see Sharkansky, 1968 vs. LeLoup and Moreland, 1978). This difference in interpretation might be traced to the preoccupation of incremen­ talists with outcomes. Such a focus ignores variations in the request strategies and the internal bargaining of the agencies. LeLoup and Moreland suggest the roles of agencies, the chief executive and legislative bodies should be redefined given what is understood about agency assertiveness and executive branch support. Agencies should not be expected to pursue a strategy of moderation in their requests. Instead, assertiveness behavior is evident throughout 25 the process as agencies seek to have executive cuts restored by the legislative body. The principal obstacle to program growth appears to be the Office of Management and Budget. In no cases in their study were program increases prepared through the initiation of OMB. acted as the incrementalists suggest: Congress large requests are cut more severely but assertive behavior resulted in greater absolute growth over the long term. There is also a general willingness to restore cuts made elsewhere in the process. This tendency explained most large funding and program changes.^® In 1976, Ronald Hrebrenar published the results of a study he undertook in the State of Washington to verify the perception that budget officials use roles as calculating devices to simplify budget decision making. "The roles fit in with one another and set up a stable pattern of mutual expectations, which do a great deal to reduce the burden of calculations for the participants."^ Hrebrenar attempts to quantitatively evaluate the importance of role perceptions as aids to budget decision making. He found, however, that the roles of the chief executive and legislature are so fragmented that no consistent pattern in behavior is apparent. He feels that the "expected behavior" of the officials simplifies decisions. "The 'calculating mechanisms’ and the 'stable pattern of mutual expectations’ are not based on role choices or cognitions for the political actors in the State of Washington, but merely as past records of budgetary behavior."2® The role variable is still very important in the budgetary process as a determinant of final budgetary outcomes. An agency is never assured of success, but the support of the governor is crucial if its budget is to be expanded. Sharkansky found 26 in his 1968 study that administrative agencies and the governor play more consistent roles than the legislature. appeared to be very important considered by the legislature. The governor’s support for an agency's request as it was There is reason to believe that the legislature will respond more readily to a request if it already has the support of the Governor than it will to a general acquisitive strategy. While such strategies are not given kind treatment by legislatures, it appears to be very important for those agencies which hope to signifi­ cantly increase the level of support for its high priority programs. Gubernatorial support is still more important than acquisitive strategies in explanations of agency success. P 11 Budget Theory and Reduction Decision Making Just as Davis suggests, there is reason to believe that incremental models for budget outcomes describe legislative decision making more accurately than decisions made in the executive branch. Advocates of the incremental or process models make valid contributions and tend to compensate for major weaknesses the other model might possess. Guhde and Mustapha feel that process oriented research can help overcome difficulties incrementalists have in explaining non-incremental adjustPP ments. c These gaps in the incrementalist explanation can be supplemented by a better understanding of role, environmental and institutional variables. The fact that budgeting is both centralized and decentralized presents problems for budget theorists. The detail and justifications for requests are prepared by program managers within guidelines provided by the central budget staff. Overall allocations are made by the 27 central authority. Agency allocations are guided by expected levels of revenue and political marginal changes. same review. interests. Most analysis is concentrated on Continuing program needs are not subjected to the The hierarchy of the bureaucracy establishes priorities, revenue levels, and restrains agency fiscal practices. There is reason to believe that decisions tend to be more centralized during times of budget reductions. As the number ofgovernmentprograms is reduced, becomes more important to insure that thoseprograms whichremain it are in fact those which the Governor and the legislature consider to be the most important. The task becomes to enforce intent while militating against those interests in the agencies which may not coincide with the desires of the leadership. Decisions made on levels of staffing and financial support come out of a complex interaction of a large group of governmental officials. Michigan, In the following groups are involved in these decisions, but their roles and influence can vary depending on the stage of the budget cycle. 1. Program Managers 2. Divisions and Bureau Chiefs 3. Agency Budget Staffs 4. Department Directors 5. Budget Examiners in the Department of Management and Budget (DKB) 6. The Director of DMB 7. The Executive Office Staff of the Governor 8. Governor 9. Appropriations Committee, and 10. The Legislative Leadership 28 The literature suggests that the three models which have evolved out of the budget theory literature apply to different aspects of the budget process. The rational/comprehensive is more feasible at the lowest decision levels (micro) and the incremental is the most evident in legislative settings (Davis, et. al., 1967). The process model transcends the range in between. Possible Seduction Decision Patterns Based on informal discussions with budget analysts in the Department of Management and Budget and in the Department of Natural Resources during the pretest of the interview questions, a mix. of rational, process and incremental behavior patterns was anticipated. The conventional wisdom of veteran analysts suggested that departments with rigid hierarchic structures and decentralized budget authority would use targeting or pro rata strategies to generate required budget savings. Departments were expected to reduce programs with low depart­ mental priority (that is lower priority in the eyes of the department leadership). Within these departments, an incremental model might emerge in those cases where there was a high level of conflict. absence of this conflict, a more orderly process was In the expected particularly in specialized and technical program areas. Several budget process variables have been proposed (see Table 2.1). Natchez and Bupp (1973) suggest a measure of agency prosperity which could be used to measure program growth and decline. The equation used to generate this measure normalizes outcomes (enabling researchers to compare agencies of different sizes) produces a percentage which when added to the prosperity scores of all other agencies equals 100. This 29 Table 2.1 Alternative Operational Measures I. Department Characteristics (LeLoup, 1973) A. B. C. II. Age: The number of years since an agency was created Size: (1) Total Operational Budget (2) Total Permanent Staff (Full Time Equated Positions) Hierarchy: Number of Executives (Grades X and above) Total FTE positions Department Assertiveness 4 Prosperity A. Assertiveness: B. Prosperity: percent change in requests to DMB by a department from the previous year's request (Sharkansky, 1968) 100 x Department Proportion Observed Year Department Mean Proportion All Observed Years (Natchez and Bupp, 1973) III. Budget Outcomes A. IV. Percent appropriations changes from previous year's appropriation Legislative Support A. Support: Mean score calculated from house votes on appropriations bill — sum of scores total number of legislators yes absent no VI. 3 2 1 Department Reduction Strategies A. Strategy: (1) Across the Board (Fair Share) (2) Priority Program (3) Discretionary 30 makes it possible to determine an agency's prosperity relative to other agencies. Sharkansky (1968) uses assertiveness scores to analyze annual appropriations. This indicator measures the tendency of departments to ask for expansions in their base programs. He found that assertive agencies enjoyed the greatest long term growth in contrast with less aggressive departments. Levels of executive support (executive recommendation vs. base year appropriation) in prior years influence agency assertiveness. The higher the level of executive support the more likely agencies will take assertive action (LeLoup, 1973). Assertive agencies tend to have less budget success than non-assertive agencies in the short term. Analyses of reduction behavior might also use measures of hierarchy, age, executive and legislative support, and revenue conserva­ tion. It is premature to collect and summarize these data without first describing a general model of the reduction process. Agency assertive­ ness and prosperity measures are included in the department summaries in Appendix E, but no predictions have been made relationships of these measures to reduction outcomes. regarding the The sample size is too small (n=11) and there is no control for changes in other, more influential variables (decision environments and reduction strategies). Once a model of budget cutting is established, tests might be applied to measure the strengths of associations among these variables. Suanary If the rational/comprehensive model fits the reduction decision environments in Michigan during 1979-1983, budget decisions should 31 demonstrate the following pattern. 1. Priorities should be used. ordering programs. This might be evident in an Decisions are made by the executive should show a distribution of cuts among departments which implement programs not considered priority concerns. These decisions will not be based on a fair share distribution of cuts. 2. A program format for allocating an agency's budget among divisions may be used. This structure helps balance inputs (staff, contracts, supply and travel costs) within totals. At lower levels of the organization, these allocation decisions can be based on fair share without effecting a department wide program reduction strategy. 3. Reduction decision will be based in part on the opportunity costs of various alternatives. Some agencies or program areas will grow only if lower priority programs pay disproportionally. This pattern is expected to be most prevalent on a departmental level but it will not be surprising to see this occurring within departments among divisions equally dependent on General Fund dollars. 4. Reductions in specific programs will be made after a careful analysis of impacts, program efficiency and public benefits. 5. Fiscal analysts and budget officials will ask for and use program performance information as reduction decisions are agreed to. If the Process Model is the most appropriate: 1. Decisions will change as environmental and situational variables shift in importance. Agencies which have grown the most rapidly in the past will be cut more. If much conflict exists between the legislature and the Governor, those depart­ ments which enjoyed the most executive support in the past will be saddled with smaller reductions. 2. Behavior patterns will be a function of the perceived roles and responsibilities of involved budget officials. 3. Decision making will be sequential. Each official modifies or builds upon decision actions at an earlier step. If the Incremental Model holds: 1. Total reductions will be distributed among all departments based on their share of the Statewide General Fund-General Purpose appropriation. Reduction targets will be evident. 32 2. A priority process will not be evident and there should be little planning focus. 3. Marginal adjustments and tradeoff strategies will be used to generate required savings within program units. 4. Current year fundings level decisions will serve as the basis for reduction decisions. Past budget decisions will be the independent variables. Officials will not re-address issues resolved in prior budget years if they can help it. 5. Programs will be eliminated in stages, that is progressively. Decision makers will be reluctant to hurt anyone's interest unnecessarily. This chapter has reviewed pertinent selections from the budget literature for a theoretical basis for analyzing budget reduction practices in Michigan. this literature, Of the three models which have evolved out of the rational/comprehensive is more feasible at the lowest decision levels (micro) and the incremental is the most evident in legislative settings (Davis, et. al., 1967). The process model applies to the range in between. Since most reduction decisions are proposed by the executive branch out of the recommendations of agency directors, it might be suggested that targets for budget cuts may be proposed through a rational/comprehensive process, communicated and packaged through a process model, and summarized and reviewed in a incremental context by the legislature. If this sequence is true, the decisions made at the agency level may prove to be the most significant — policy content — have the greatest and have a greater impact on budget cut decisions than either the Governor or the legislative leadership. This study analyzes the Michigan experience and attempts to illustrate how reduction issues were resolved and to explain why these cuts were made as they were. 33 Notes 1. Herbert Simon, Administrative Behavior. (New York, NY, 1957). 2. V.O. Key, "The Lack of Budgetary Theory." Science Review. Vol. 34 (1940), p. 1143. 3. Frank Fisher, Politics Values and Public Policy. (Boulder, CO, 1980) 4. Charles E. Lindblom, "Decision Making in Taxation and Expenditures," in Golembiewski, Robert T. and Jack Rubin, eds. Public Budgeting and EAnanaa; Readinga__in-Theory and Practice. (Haska, Illinois, 1975), p. 289. 5. Herbert A. Simon, op cit pp. 38-39. 6. Joint Economic Committee Hearings, The Planning-Programming S u t e U n K ,?ygtaa; Progress and Potentials. (Washington, D.C. 1967), pp. 207, 210. 7. Charles E. Lindblom, "The Science of Muddling Through" Administration Review. Vol. 19 (1959), p. 80. 8. Aaron Wildavsky, The Politics of the Budgetary Process. (Boston, MA, 1964), p. 15. 9. Richard F. Fenno, Jr., The Power of the Purse. (Boston, MA, 1966), Otto A. Davis, M.A.H. Dempster and Aaron Wildavsky, "A Theory of the Budgetary Process." The American Political Science Quarterly. Vol. 60 (September, 1966), pp. 529-547. 10. Wildavsky, Qp.....gj.fr,,, p. vi. American Political Public 11. Ibid. p. 1983. 12. Joint Economic Committee, o p . cit.. pp 206-214. 13. Peter B. Natchez and Irvin C. Bupp, "Policy and Priority in the Budgetary Process." The American Political Science Review. Vol. 67 (1973), P. 963. 14. Simon, Qp,.,gj.fr,, p. xxx - xxxii. 15. See John Gist, "Increment and Base in the Congressional Appropriations Process." American.^Journal,__of Political Science. Vol. 21 (May, 1977), PP. 341-352. 16. Allen Schick, ed. D.C. (1980), p. 158. 17. Wildavsky, op. cit.. p. 21. Perspectives on Budgeting. Washington, 3M 18. 19. (LeLoup & Moreland, p. 189). Wildavsky, o p . oit.. pp. 160-161. 20. Ronald Hrebrenar, "The Utility of Role in Budget Decision Making," Western Political Quarterly. Vol. 29 (December, 1976), p. 588. 21. Sharkanski, 1968, p. 1230. 22. Robert Ghude and Husain Mustapha, "Budget Making in Ohio: A Test of the Process Model," Western Political Quarterly. Vol. 34, (December, 1981), p. 592. 23. Natchez and Bupp. o p .cit.. p. 959. CHAPTER III Methodology Introduction In 1964, Aaron Wildavsky observed that the way that budget decisions are made will determine what decisions will be made. process followed, the information generated and the timetables used establish a decision framework. officials The The roles and relationships of budget influence the political interaction and compromises which result in these allocation decisions. From this, it might be reasoned that it is important to have an understanding of the decision process if one is to project the results of these decisions. If government officials are contemplating major budget cut decisions, it is important to be confident that the process selected will facilitate the desired results. Based on the literature reviewed, it appears that considerable overlap exists especially in matters relating to the ways that budget officials reach funding decisions. All three models address the process followed to reach agreement on funding levels. There is considerable disagreement however on normative grounds and over the value of each model as an explanation of the process of budget decision making. The purpose of this research is to determine if one budget model best 35 describes the process followed in Michigan as budget cut decisions were made. Hypothesis Since most reduction decisions are proposed by the executive branch out of the recommendations of agency directors, it is suggested that targets for budget cuts may be proposed through a centralized incremental process, analyzed in a rational/comprehensive environment, communicated and packaged through a process model, and summarized and reviewed in an incremental context by the legislature. If this sequence is true and if the decisions made at the agency level are evident in final budget cuts, the incremental and process explanations may offer the best basis for predicting final budget outcomes. This dissertation is both diagnostic research questions are addressed; "How and are descriptive.Two basic the majorityof budget reduction decisions made in state governments?", and "what budget model best describes the process followed in Michigan between 1979 and 1983?" Answers to these questions are developed through a case study. Studies such as this one are useful to theorists who seek to define variable units, to propose relationships among these units, and to define the system boundaries that encompass budget reduction decisions. These issues can be resolved only after these units are defined, measures are developed and the strength of relationships among these variables are determined. This study attempts to provide a basis for predictions of how reduction decisions will be made, expected. and what outcomesmight be This information can be used to assure an open and fair exchange as final reduction targets are agreed upon. 37 Data have been gathered through personal interviews and a survey of thirty-two key budget officials. Both employ open-ended questions and are designed to provide insights into changes that may have occurred in the ways that reduction decisions were made between 1979 and 1983 (see Appendix C). Under ideal conditions, this research would have gathered information from no fewer than thirty-eight states (seventy-five percent sample). Each data set would have included reduction outcomes in representative classes of agencies (based on age, size, mission and political sensitivity), and would have covered a range of possible fiscal crises extending from minor or technical adjustments to major cutback decisions. These states would have had similar budgetary histories and stable decision making environments. Such a study is not feasible or possible given the scale of research necessary and the dissimilarity of state governments. Furthermore, the lack of any theoretical work in this area makes data collection difficult. No evaluative units have been proposed which could be used to test variable relationships. If reduction strategies could be expressed as evaluative units, the consequences predicted. of a given strategy or budgetary outcome could be These could then be compared with actual reductions and modifications to general budget propositions might be suggested. The difficulty with this research design is that it cannot be applied to the Michigan case. ment. There are nineteen departments in Michigan State Govern­ As time passed reduction strategies changed. occurred at unpredictable points in the process. These shifts Interviews offered some explanations but final reduction decisions could not be related to these changes. If this information were available, propositions about 38 the relationships among these units or reduction variables could have been proposed. These might have included categorical, sequential, or determinate (direct or inverse) relationships. It is this interaction among variables that is the most interesting aspect of studies such as this one. Unfortunately, the paucity of information makes such analysis impracticable. Michigan Case Study Very few state governments have been faced with major cutback decisions for more than two successive years. It is also difficult to identify a sample of states which have faced similar economic circum­ stances and comparable revenue crises. Of those states which have experienced severe budget crises (Massachusetts, Michigan, California, Utah, Ohio and Washington) Michigan was selected as a representative case. It provided an opportunity to observe the interaction of a wide range of variables, and the crisis lasted more than two years. was also selected because the author worked there. Michigan This reduced the probable costs of doing this study in another state government. This study will review budget decisions made Government of Michigan between 1975 and 1983. in the State Michigan experienced the most severe and extended financial crisis of all states in the upper midwest. The executive leadership was unchanged from the onset of the crisis in 1975 through 1982. This thereby reduced the possible effect changes in political philosophy might have had on reduction strategies and agency decisions. The State Government of Michigan was unable to influence economic variables which caused the steady deterioration of revenue during this ten-year period. Like other governments, Michigan 39 is constitutionally mandated to maintain a balanced budget. It has a track record of budget innovation and the tools available to government officials are similar to those of other states (the line-item veto, variable budget structures and a progressive accounting system). Research Design Information used to test this hypothesis was collected through a survey and interviews with agency officials (7), fiscal analysts (13) and key budget officials (7). closed and open ended. The interview and survey questions were The closed questions offered an opportunity to compare perceptions of these groups while open ended questions were included to give respondents an opportunity to give their personal perceptions and explanations of the way reduction decisions were reached. Four aspects of these decisions were explored: the strategies used by agency and key budget officials to generate the required savings; the decision units used in reduction negotiations; the infor­ mation used by budget officials and analysts; and the relative influence of agency officials, program managers, fiscal analysts and key budget officials in reduction negotiations. The response rate to the written budget officer survey was fairly good. Of the twelve surveys sent, seven were completed and returned (one unsealed envelope was delivered without an enclosed form). The interviews were tape recorded and several also answered the closed questions on the forms themselves. A copy of the questions asked in the course of this phase of the study may be found in Appendix A. These questions were pretested by budget analysts in the Department of Management and Budget for the 40 purpose of improving their comprehensiveness, emphasis and appropri­ ateness of items included in the interview. Some modifications were made to focus closer on the official's predispositions and to draw out the nuances of budget decisions which might be evident in responses. The scope, both in time and detail, was restricted and an attempt was made to produce information useful to these officials which meeting the data requirements of the study. Three patterns of budget decision making have been proposed (See Table 3.1). If the Michigan case follows one of these general decision making patterns, then it might be reasoned that that theory might serve as an appropriate model for the reduction process. The questionnaire developed for this study included twenty-one open ended questions. Each question was designed to provide insight into the thought processes of budget officials and to record their views on their role in the reduction process. Some questions provided background for more than one dimension of the reduction problem. These responses were reviewed with several evaluation standards in mind. In each case, the responses were analyzed and compared to key or critical answers which could indicate which model of the budget process might best describe the observed behavior. The standards, focus and the critical responses used to review these responses are summarized in Table 3.2. information have also been made available. Other sources of Data were gathered from department management plans, appropriation acts, executive orders, and reports prepared by DMB and the fiscal agencies. 41 Table 3.1 Alternative Budget Reduction Models Rational/ Comprehensive Model Process Model Incremental Model Reduce by Program Progressive Cuts Program Cuts Fair Share & Progressive Cuts By Program Program Elements vs Funding Levels Current Year Funding Level Current Year + Growth History Long Term Strategies Short Term Issues Incremental Decisions From State to Stage Incremental Adjustments Budget Official Influence Priorities Used Priorities of Dominant Official Given Stage of Process No Priorities— Only Influence Dominant Decision Rule Program Efficiency & Effectiveness Situations, Roles, & Decision making Environments Dominate Elected Official Decide, Others Recommend Assignment Opportunity Costs Important Component Opportunity Costs May be Taken Into Account Opportunity Costs Rarely Considered Reduction Issue Strategy Decision Dnits Issues Development Table 3.2 Evaluation Framework Dimension Strategy Evaluation Standard Goal Attainment Focus Reduction Plans Critical Responses • • • • Target by Program Progressive Program Cuts Pro R ata Assignment of Reduction Targets Incremental Reduction Decision Units Language of Reductions Fiscal Detail • • • • Organizational Units Appropriations Level Program Structure Items of Expense Relevant Information Reduction Criteria Program Information • • • • Priority Issues Program Efficiency Program Services Planning Issues and Opportunities Budget Official Influence Institutional and Individual Values Reduction Recom mendations • • Bargaining Organizational Priorities, Goals, and Objectives Personal Priorities Influence Decisions Through Selective Information • • Table 3.2 (Continued) Evaluation Framework Dimension Evaluation Standard Criteria for Assignment of Reductions Focus Critical Responses Agency Impact Statem ents • • Executive Order Reductions Constituency Activity Dominance Final Reduction Order Programs, Departments and Service Areas Assignment Final Reduction Order Assignment of Cuts Decision Environment Program Efficiency Responsibility of Budget Official Information Used to Make Reduction Decisions Assignment Changes Made in Successive Stages of the Process Evaluation of Results The interviews and survey questions were designed to provide information on three aspects of reduction decision making. Six questions asked for descriptions of the strategies used in the agencies in response to DMB. Nine questions addressed the framework or decision units used to evaluate reduction plans. Ten questions focused on the information provided and used by those who made reduction decisions. The relative influence of participants in these decisions was addressed directly and indirectly by four open ended and six closed questions. Many of the open questions provided information which could be used in more than one aspect or dimension. The results were tabulated on a form like that shown in Table 3.3. Sfaiafcggy These six questions were designed to determine if department plans reflected a program, target or a changing reduction strategy. A program focus might indicate a rational approach, a target strategy reflects an incremental bias, and changing strategies might suggest that external factors or variables influenced the choice of technique. Decision Units and Relevant Information These questions reduction hearings. suggested how reductions were discussed in Were they structured around an organization, program, or appropriation unit structure? How important or useful was information on past appropriation levels and patterns of expenditure? Were department priorities presented or used? performance information? Were How important was program planning issues or opportunities 45 Table 3.3 Interview Results Worksheet Dimension Strategy: Reduce by Program Reduce by Share Changing Strategies with Time Decision Units: Organization Structure Program Structure Appropriation Unit Structure Items of Expense Relevant Information: Appropriation History Appropriation Level Department Priorities Program Performance Issue and Opportunities Influence - Who and When: Department Officials Agency Program Manager Fiscal Analysts State Budget Officials Interview Group: Agency Budget Offices (7) Fiscal Analysts (13) State Budget Officials (7) H6 discussed in agency impact narratives? Responses to these questions assisted the researcher in analyzing the basis of these discussions and bias of budget officials. Much of the information was generated through the open ended questions. The researcher was looking for patterns of influence by agency officials, the Department of Management and Budget, fiscal analysts and state budget officials (State Budget Director and the Legislative Quadrant). Were different groups influential at different stages in the process? Who could make changes or recommendations and when could they make the greatest impact on these decisions? The techniques used to influence were noted when offered by interviewees, but this information was not always volunteered. The oral interviews proved to be the most useful from a descriptive point of view. Representative perceptions from each of the three groups provided important insights into the dynamics of the process. This information has been presented in the fifth chapter and helps to interpret the results tallied from the written responses to the closed questions and the survey. Limitations of this Research Design The one-shot case study is the simplest of all research designs. Information is collected after the activity has been in place for some time on the impacts or consequences arising from the observed behavior. Of all research designs used in evaluation research, the case study is the weakest. Results have not been based on any comparisons with other state practices, and few of the individuals interviewed were unaffected by the fiscal crisis. It is assumed that budget decisions were made in the same way in 1980-83 as they were in 1972-73 or 1984-85. Since it is difficult to validate this conclusion, this research is unable to take into account other explanations for any changes in administrative behavior. CHAPTER IV MICHIGAN 1979-1983 Overview The budget process used in Michigan is both fairly straight forward and highly complex. Like the cycles of other states and local units of government, there are essentially three phases in the Michigan budget process: Request Phase, Legislative Phase and Allotment Phase (See Table 4.1). The Executive Budget Recommendation of the Governor is the principal product of the request phase. The result of the legislative part of the cycle is a series of enrolled bills presented for line item veto and/or signature to the Governor. or expenditure phase. The final phase is the allotment The typical budget process begins nearly 18 months before the allotment phase begins. The groups are involved in the budget process may be involved in one or more phases of the budget cycle. Budget officials are usually directly involved in only one of the phases, but agency budget personnel and budget analysts in the Bureau of the Budget are active in all three phases. Responsibility is generally limited to one group but all have important roles to fill. These roles often change as one phase is completed and the next one begins. Table 4.2. These changes are illustrated in The roles identified are used by Hrebenar and Howard as they 48 Table 4.1 Budget Process Participants Request Phase PartJciDant Grouo Executive: Governor & Staff Bureau of the Budget DMB Accounting Agency Directors & Staff Eureaus, Divisions & Staff (x) = Allotment _ Phase X X X X X X X X X X (x) X (x) X (x) X Legislative: Legislators Appropriations Committees Fiscal Agencies Note: Legislative Phase X Information only Table 4.2 Budget Process Roles ParticiDant Grouo Governor Gov. Staff Bureau of the Budget Agency Directors Agency Budget Officers Program Managers Bureau Clients Legislators Appropriations Committees Fiscal Agencies Request Phase Legislative Phase .. Revenue Balancer Planner Prloritizer Budget Reviewer Planner/Rev. Bal. Budget Reviewer Rev. Bal./Budget Reviewer Program Advocate Planner Prloritizer Budget Maker (P) Revenue Balancer -----Prloritizer (P) Budget Reviewer (P) Planner Balancer Program Advocate Planner Revenue Balancer Budget Rev. (P) Controller Program Advocate Program Advocate Manager Program Advocate Planner _ _ Budget Review (P) Note: (P) a Passive role in most cases. Budget Reviewer Planner Prloritizer Budget Reviewer Allotment Phase ------ -- 50 attempted to offer explanations for different impact officials might have on budget decisions. suggests. The system is not as neat as Table H.2 Hrebrenar found that it was not unusual to find individuals acting out one role in an appropriations committee meeting one day (e.g. planner/ prloritizer) and the next day acting as a revenue balancer on one issue, a program advocate on another, and a budget reviewer (concentrating on detail) at another time. The expected role of the official dominating the next stage is more important than the role actually exercised. Some questions might be left unresolved in deference to the officials higher in the responsi­ bility hierarchy. This expectation of "role" is discussed by Wildavsky facilitates decisions on an agency’s budget levels (position and dollars). As successive stages in a given phase are completed, the number of issues or decisions needed is reduced. By the time the Governor’s staff is involved at the end of the executive phase, there might be as few as two or three major policy issues or funding alternatives which could not be agreed to by agency directors and the State Budget Director (the Director of the Department of Management and Budget). As decision makers higher in the decision making hierarchy become involved, these issues typically become more broad in scope and fiscal details are minimized. These issues usually deal with alternative financing arrangements, total allocations for individual departments, and the revenue implications if all departments are funded at recommended levels. Generally, revenue discussions are restricted to the General Fund - General Purpose side of the budget. As the later stages of a given phase are reached, the revenue balancer role 51 prevails. This progression of dominant roles is illustrated in Figure 4.1. The budget process in 1978 was not much different from that followed in other budget years. All participants were conditioned to expect certain behavior and roles of each other. Most analysis was directed toward the marginal or incremental requests proposed by the departments and recommended by the decision makers active in the agency phase. No conscious effort to review the base has been evident. While it was acknowledged that long range planning is rarely asked for, few officials feel that this has created any serious problems or impediments to rational budget analysis. Back in 1973, an organized approach to comprehensive planning was proposed and withdrawn very quickly the following year. While the decision to merge planning functions into other bureaus of the Executive Office grew out of manage­ ment problems within the new Planning Division, the germ survived for a short time in the Bureau of the Budget. Between 1973 and Spring 1975, this division (Planning and Policy Analysis Division) was responsible for policy analysis and the development of issues to be included in the annually published Program Policy Guidelines. This document was to be the "planning1* front-end for the Executive Budget. Directions were to have been given to the various agencies of the Executive Branch to address emerging issues and priorities for inclusion in the Governor's budget recommendations. In a contemporary analysis of the PBES budget program, initiated by they Legislative House Fiscal Agency, the program budget approach (rational/comprehensive budget model) was generally applauded but the Executive Branch was criticized for failing to fully implement the Figure 4.1 Role Premises Michigan Executive Budget Process Stage of ttie Process_____ I. Uai:inunt Characters Issue Focus State Budget Director 1. 2. Executive Phase: Governor's Pol icy Qiidelines Policy Initiatives Anticipated Revenues Planner/Prior it izer Continuing nudget Needs Department Budget Officer Funding Needs for Department Programs Budget Reviewer Progran Revisions Department Director Policy Initiatives Planner/Priori t izer IWmugenent Plan 1. Deportment Budget Officer 1. 2. Budget Reviewer Revenue Balancer 2. Department Director Policy initiatives Program Priorities 3. 1MB Budget Analyst 1. 2. 3. M. Role of Decisionnakcr Policy Initiatives Special Fund Revenue Est. & Budget Policy Program Efficiency & El feet iveness Alternative Sources of Funding Planner/Prior itizer Budget Reviewer Program Analyst Revenue Balancer 4. State Budget Director Management Plan Issues Budget Reviewer Revenue Balancer 5. Governor & Staff Assistants 1. 2. 3. Policy Initiatives General Fund Revenue Est. Manngunant Plan Issues Planner/Priori t izer Revenue Balancer Budget Reviewer 1. E x e c u t i v e Budget Recannendalions Budget Reviewer* Planner 2. Revenue E s t i m a t e s Revenue B a l a n c e r 1. 2. P r o g r a m Mi ndi ng L e v e l s (Jvpiirtinent I s s u e s Budget R e v i e we r Budget Rcviev. or Legislative Pliuse: Cfcnmittee Orientation Siihccnmi t t e e H e a r i n g s Pi s e a l Agency S t u f f i h u i j c a i m i t t e e Cha i r men Figure 4.1 (. Continued ) Dominant Characters Issue Focus 2. Department Director Program Priorities Governor's Advocate 3. Department Budget Officer Departnent Program Needs Department Advocate 4. EM3 Budget Analyst Executive Budget Kecomnendation Governor's Advocate 5. Fiscal Agency Analyst Department Request & Executive Budget Reccmnendation Budget Reviewer 1. Department Director Program Priorities Governor's Advocate 2. Subccrmiittee Chuinnan Subcommittee llecaimendations Ctximittee Advocate 3. Appropriations Qxnnittee Ch. 1. Legislative Priorities Budget Reviewer 2. Department Appropriation Revenue Balancer Qxnnittee Information Needs Revenue Est m a t e s Budget Reviewer Revenue Analyst Stage of the -•••Process ■ Subconmittee Hearings Appropriations Qxnnittee Conference Qxnnittee Role of Decisiorinaker 4. Fiscal Agency Analyst 1. 2. 5. QIB Budget Analyst & State Budget Director Department Appropriations Governor's Advocate Revenue Balancer 1. Oonf. Qxnnittee Q m i m i i n 1. 2. Department Issues Itouse-Senate Differences Planner/Priori t izer Planner/Priori tizer 2. Fiscal Agency Analyst 1. 2. 3. Qxnnittee Infomation Needs Revenue Estimates Department Needs Budget Reviewer Revenue Analyst Departnent Advocate (in its absence 1. Governor and Staff 1. 2. Revenue Estinutes IV)1 icy Rccumendut ions Revenue Balancer Governor's Advocate 2. IMi Budget Analyst 1. 2. I.ine-item Appropriation Itecannend l.ine-itan Vetoes Budget Reviewer Governor's .Advocate 3. S t a t e Budget Pi r e c t o r Vein Recrmucmliit i n n s Allotment Pliuse: .Vetoes Re venu e l i n l a n e e r 4 Plu n n er/P rio ri I izer Planner 4 I-'igure 4.1 ( Continued ) Dominant Cha ractera Stage of the ■■ Process III. Issue Focus Hole of Pecisionmker Allotment Phase: ( Continued ) Allotments 1. Division Chief ( ProgramManager ) planned Expenditures Planner/Prioritizer 2. Department Budget Officer 1. Planned Expenditures 2. Budget Detail Budget Reviewer Revenue Balancer 55 system. The PBES system failed to effect budgetary outcomes largely because of:^ 1. Lack of credibility of the Executive Revenue Forecasting Program, i.e., timeliness, legislative input, numerous interim revenue changes lacking basic economic support in justifi­ cation of said changes. 2. Executive failure to develop a formal 'planning1 program — short to long range — incorporating public policy. 3. Failure of the executive to develop effective data processing capability correlated with the state fiscal-budgetary system. 4. Failure to assert strong executive authority to assure fulfillment of prescribed system commitments. 5. Departmental failure to adhere to system-related executive directives and requirements regarding goals, objectives, evaluation, effectiveness. 6. Failure of state departments, bureaus and agencies to develop program measurements and related program data. 7. Willingness of the executive to allow the program budget document to return to a ’conventional' incremental budget. The 1976-77 executive budget is void of program information — goals, objectives and evaluation criteria. There is no visible linkage to state policy; no visible linkage to a formal program plan; no visible linkage of program funding to program performance; no visible reflection of program outputs based on funding levels or corresponding priority needs to achieve state goals or objectives. The Michigan program budgeting system embodied many of the same components and documentation evident in program budgets prepared by the federal and other state governments. Designed and implemented through the consulting efforts of Dr. Robert J. Mowitz of the University of Pennsylvania, program the PBES budget was to add provided information and performance analysis (program impact efficiency and effectiveness) and to establish a rational basis for budget decisions. The Michigan experiment with program budgeting ended in 1974 in the words of the "green budget", the Executive Recommendations for 1975-76. Significant because of what is not said, the introduction to this budget described the content of the budget document. The program structure used in the three earlier budgets was abandoned in favor of the appropriation limit structure used by the legislature, and the Governor admitted that program performance information was only marginally better than that information used prior to implementation of the PBES budget system. "Because of the highly uneven nature of the key resource and performance data across agency programs, they have been generally deleted from this presentation. It is the expectation of the executive budget system to produce program data in the context of established objectives in a document apart from this detail document and at a time which precedes budget detail development both by the Governor and the legislature. The first such document is planned for early fall 1975."2 The document referred to was to be titled the Michigan Program Plan and was to present the responsibilities, management goals and objectives, and program activities (and the referenced performance measures) of each department in the Executive Branch. The document never progressed beyond the design stage however and the promised objective-based program plan never materialized. This could be attributed more to changes which grew out of a general reorganization of the Bureau of the Budget and the dissolution of the Planning and Policy Analysis Division than to any basic flaw in the concept, DMB policy analysts suggest. The shift in priority from "crisis mongering" to enhanced management control and improved budget information paralleled the change from the planning orientation of PBES to the management focus evident in the "Target Budget" approach which followed. The 1975-76 Executive Budget thus represents a high point of Michigan's rational budget systems, and marks the end of a period of experimentation and innovation in Michigan budget making that began during the closing days of the Romney Administration in 1968. The Legacy of the Program Budget Evaluation System Program performance was emphasized and Legislative awareness of the systems concepts of PPB was increased. improved, Program information was and agencies have been made more aware of the benefits realized through improved management and planning programs. These benefits did not outweigh other problems. to the way that the system was implemented. Some can be traced Others represent more serious flaws which could not be corrected easily.According to Allen Schick, the Federal experiment with program budgeting failedbecause: 1. The connection between program analysis and budgeting was not appreciated. 2. PPB was applied across the board by people insensitive to the way budget decisions had been made. 3. Inadequate resources were committed to the job. 4. PPB worked against the bargaining-incremental mode of budgeting. This mode gave expression and representation to diverse political interests. This might have been neglected if decisions had been made by PPB analysts. 5. Information requirements of the budget process were not appreciated. 6. PPB failed to integrate programs which crossed department lines. 7. Budget and appropriation accounts could not easily be "crosswalked" into the PPB program structure. 8. PPB was conceived from an Executive perspective. take the role played by Congress into account.3 It did not In 197^» an unpublished Bureau of the Budget internal audit of the performance of Michigan's Program Budget Evaluation System uncovered other difficulties. 58 1. There was no planning perspective. There were no projections of future benefits or of possible changes in needs or demands for state services. 2. A common framework for executive and legislative use was not attainable. 3. Department budget requests lacked an analytic basis for review or approval. ij. Program objectives were poorly specified. 5. Budget policy was not sensitive to revenue constraints. 6. PPB was issued from the Governor's Office with little or no involvement from the Legislature. This brought increased conflict, impaired information flow, and further confused decisions made in appropriations hearings. 7. Program information was poor and implications of target responses could not be easily defined. 8. Analytical output was poor, both in quality and quantity especially for requests for program revision and expansions. 9. Analysts could not adapt budget to a rapidly changing revenue picture. No rational method was available to reduce expenditures. In short, program budgeting failed to effect how budget decisions were made. This failure resulted from implementation problems and from basic problems with the theory behind this closed system process. 1. PPB failed to capture the contingencies of decision making. 2. PPB assumed that program manager have control over the variables affecting goal attainment. However, public programs are not closed systems. Decision makers do not have control over their task environments. Therefore, they can not be held accountable for non-attainment of program objectives. The role of administrators was in fact too strong for the routines, forms, doctrines of PPB to penetrate. decision making is a hollow one. wanted was not the The objective of rational What proponents might really have orderly evaluation of alternatives, but a transparency in budget decisions to enable budget officials to insure 59 that budget policy was enforced and pursued in daily administrative decisions. The most recent effort to establish planning objectives in budgeting came from‘the House Fiscal Agency in 1976. This proposal to reestablish the program budget concept for legislative decision making included a plan for the reorganization of the appropriations structure and reducing the number of deliberative committees. It also proposed that budget analysis be centralized in a single House Budget Committee which could draw on the expertise of the standing committees for fiscal decision making. The legislative process was also to be revised; patterned after the changes made by the US Congress in 1971*. Budget decisions were to be centralized and other standing committees were to play a more active role in priority decision making. During the years immediately before this proposal was presented to the House Leadership, the House Fiscal Agency established an in-house program evaluation staff made up of refugees from the Department of Management and Budget (nicknamed the "leper unit" this group was called the Legislative Program Evaluation and Review Dnit — LEPR). Not long after this proposal surfaced, the evaluation unit was disbanded and the Director of the agency was encouraged to find work elsewhere. One of the proposals which survived the fiscal agency reorganization was the expansion of the revenue analysis expertise in fiscal agencies. The Senate Fiscal Agency, like the House Fiscal Agency, devoted more effort and developed credibility as a co-equal revenue and economic analysis organization to the Office of Revenue and Tax Analysis in DMB. 60 The legacies of program budgeting remained after the demise of PBES and the short lived legislative experiment with policy planning. A cadre of budget officials had been trained in principles of program budgeting and management by objectives and by 1978, had assumed positions of authority in DMB, in the executive departments and in the fiscal agencies. The trappings of program budgeting had been set aside, but a review of agency budget practice in 1985 reveals a continued emphasis on program objectives, impacts, program outputs and on service needs and demands. These concepts have been applied in such a way that the comprehensive system of accounts, cost centers, and objects of expenditure are enveloped in a program framework not unlike the program structure that dominated the PBES system. All reflect the management needs of the agency yet mesh easily into the appropriation unit structure. The majority of the analytical time devoted to the evaluation of budget requests was and continues to be spent on the changes requested by the agencies Michigan’s budget and those changes recommended by the Governor. process incremental analysis. in 1978 was dominated by marginal or Documentation focused on the variable levels of program impact and output evident in different levels of financial support (e.g. 90, 100, and 110 percent of current year funding levels). Program Revision Requests were to describe the different benefit levels which might be possible given alternative means of providing the proposed services. Base program levels were generally ignored however, and only a limited number of departments used a priority system for allocating department target levels among their program units. 61 Budget officials active in budget reduction decisions during the 1979-1983 period were trained and acclimated to Michigan budgeting during the experiments with program and target budgeting. usefulness of different aspects of these two systems The influenced budgeters in 1980 as information describing the effects or impacts of proposed cuts for Michigan government and its citizens was requested. Program budget and equimarginal concepts of these two budget processes provided a framework for information generated as reduction decisions were made. Spending Patterns During the 1970’s, the Governor appeared to recommend a level of expenditure very close to the most optimistic estimates of general fund revenue. During the 1968-1978 interval, state government spending rose from $2,398.9 million to $8,031.1 million. Significant growth occurred in two basic areas as government doubled in budget sized between 1967 and 1973 and doubled again before the onset of the 1980 recession. While all of State Government expanded at ;moderate rates during this time, the Social Services and Local Grants areas of the budget grew disproportionately at the apparent expense of Higher Education, School Aid and Transportation portions of the budget. See Table A.3 . The drop in the Health share might be attributed to the transfer of environmental health protection programs to the Department of Natural Resources in 1971. The completion of much of the Interstate highway system during the middle 1970’s might explain the drop in transportation spending. Growth in the human services areas was designed to improve the quality of life for citizens with annual incomes at or near poverty levels. 62 Table 4.3 Categorical Spending Patterns 1968-1978 Percent of Total 1067-68 1977-78 Social Services 16.9? 28.0? School Aid (K-12) 31.9 24.3 Transportation 16.1 10.0 Higher Education 10.9 8.5 General Government* 5.4 8.7 Health 7.2 6.5 Grants 4.5 7.1 Safety and Corrections 2.3 3.0 Debt Service 1.7 1.2 All Other 3.3 2.9 100.0? 100.0? Cateaorv * General Government includes Executive, Legislative and judiciary, and the departments of attorney general, civil rights, civil service, management and budget, state treasury, agriculture, commerce, labor, licensing and regulation, and natural resources. 63 This is not to say that there was a conscious decision to reduce the priorities of K-12 education. There were steady and general drops in student enrollment during this time and larger shares of the costs of education were supported through increases collections. in local property tax Government employment grew steadily through the 1974-75 recession and the tendency to expand to the limits of available revenue is evident in the trends in General Fund surplus levels during this period. The lapse normally was in the neighborhood of one percent of total revenue during good economic years (FY 77 and 78) later in the decade, but in the order of 10 percent before the general expansion of state government during the 1965-75 period. State expenditures doubled between 1967-68 and 1972-73 (from $2,423.5 million to $5,021.8 million respectively). Institutional Responses to Fiscal Stress The Executive Office of the Governor is solely responsible for the preparation and projection of estimated revenue (State Constitution of Michigan Article V, section 18). The legislature is indirectly informed through the Executive Budget Recommendations. Between 1970 and 1976 there is no evidence that the legislature was given revenue estimates before the executive budget was sent for its consideration. This should not be altogether surprising once there is no legal mandate or legislative policy which would require the Governor to include the legislature in the revenue estimation program.1* These revenue estimates have always been closely related to the level of funding required to support the executive recommendations. In its 1976 critique of the budget process, the House Fiscal Agency commented on this problem. 64 "...visible or documented estimates are not always projected with realism, but rather geared to coincide directly with the Governor's budget recommendations. In essence, the crux of the claim charge that the initial (released) annual revenue estimates related more to the level of funding needed to support the Governor’s budget programs, rather than the actual expected level of revenue receipts. The agency felt that the governor’s revenue estimates could not be relied upon. During the 1976-77 legislative phase (just after the recommendations are submitted to the legislature in January 1976), fifty change requests were sent to the legislature by the Executive Office (these amounted to $95.0 million in increased costs). changes were being submitted, While these the Executive Office also submitted changes in the revenue forecasts. In most cases, these changes demonstrated that more revenue was available than was earlier suggested and that the requested changes in the budget recommendations would be supported by the increase. During the 1976-77 budget request cycle, the Executive Recommendation grew from $3,275.0 million to an adjusted recommendation of $3»370.0 with these increases supported by comparable increases in the executive revenue estimates. to the credibility problem by The legislature responded creating its own Joint Legislative Economic and Revenue Projection Office. The leadership hoped that this check on the executive might both have the effect of improving the reliability of the executive estimates as well as providing for improved fiscal planning capabilities in the two fiscal agencies. However, changes in the composition and volatility of the national and Michigan economies created many difficulties for all revenue analysts alike. As Table 4.4 shows (Unemployment Rates) unemployment rates in Michigan during times of national recession are nearly sixty percent 65 higher than the rates for the nation as a whole. The 1979 recession, the most severe recession since the Depression, forced Michigan to diversify its economy (Table if.5). In 1960, 32.1 percent of personal income came from durable goods manufacturing and 8.2 percent in "other" manufacturing. In 1979 these percentages had decreased by 25.6 percent and 6.1 percent respectively. recession, In 1982, during the peak of the durable goods manufacturing was only 21.3 percent. One obvious sign of a continuing problem is the 1984 twelve percent unemployment rate. This only emphasizes the fact that new jobs are not being created at a fast enough rate in the service sector to make up for the permanent loss of jobs in other industries. A quick glance at Table 4.6 (General Fund-General Purpose Fund Balances 1967-1982) gives a general picture of the effects of the 1980-1981 recession on Michigan State Government. Between 1979 and mid year 1982, more than 6,000 jobs had been eliminated and $977,300.00 cut from General Fund - General Purpose programs through Executive orders. Before the crisis was over more than $1,203.1 million in general fund cuts (annual spending) had been made. To say that these actions had a severe impact on government is an understatement. Brazer suggests that these cuts were the result of a deliberate policy of political action to reduce the size of state government (Brazer, 1980). Writing in 1982, shortly after Michigan's bond rating was reduced from 'A' to 'Baa', Harvey Brazer suggested that the way Michigan State Government responded to the fiscal crisis was much different from the approaches of other administrations to earlier recessions. He observes that each of the recessions of 1949, 1954, 1958, 1961 and 1970 caused serious revenue problems. In most cases the legislature went along with 66 Table 4.4 Michigan vs. U.S. Unemployment Rates 1956 to 1984 Percent Unemployed Calendar Year Michigan Rate United States Rate 1956 1957 1958 1959 1960 6.9? 6.6 13.7 8.4 6.7 4.1? 4.3 6.8 5.5 5.5 1961 1962 1963 1964 1965 10.1 7.0 5.6 4.8 3.9 6.7 5.5 5.7 5.2 4.5 1966 1967 1968 1969 1970 3.5 4.5 4.3 4.0 6.7 3.8 3.8 3.6 3.5 4.9 1971 1972 1973 1974 1975 7.6 7.0 5.8 7.4 12.5 5.9 5.6 4.9 5.6 8.5 1976 1977 1978 1979 1980 9.4 8.2 6.9 7.8 12.4 7.7 7.1 6.1 5.8 7.1 1981 1982 1983 est 1984 est 12.4 15.5 14.1 11.1 7.6 9.7 9.6 7.9 Summary • Average » High * Low * Average Deviation • Last 10-Year Average Source: Michigan Rate 8.1? 15.5 3.5 2.6 11.0 Senate Fiscal Agency, 1984 United States ..Rate 5.9? 9.7 3.5 1.3 7.7 Difference 2.2? 5.8 -01.3 3.3 67 Table 4.5 Wage and Salary Employment by Industry in Michigan Selected Years 1960-1982 Industries 1960 •Thousands of Persons 1979 1965 1973 1982 Durable Goods, Mfg. Non-durable Goods, Mfg. Construction Tans., Comm. & Utilities 770.6 197.0 97.2 199.9 888.8 213.0 111.1 192.4 952.1 225.1 132.7 152.4 928.3 222.9 139.8 158.8 680.7 193.4 90.0 141.5 Wholesale Trade Retail Trade Finance, Real Estate & Insurance Service 102.9 347.9 119.8 390.7 150.6 503.0 172.1 585.2 149.9 539.5 82.8 263.7 94.0 326.7 130.2 489.0 153.2 631.1 151.7 652.7 Mining Federal Government State Government Local Government 15.5 46.3 71.1 215.3 13.4 50.6 83.3 261.0 12.9 55.0 123.9 355.4 13.4 56.0 150.3 416.7 10.6 57.4 144.1 377.8 2,410.2 2,745.6 3,282.3 3,627.9 3,189.3 Total Percent of Total Industries________________1S6J0_____ ,1965_____ 1973 1979_____ Durable Goods, Mfg. Non-durable Goods, Mfg. Construction Tans., Comm. & Utilities 32.1$ 8.2 4.0 8.3 32.5$ 7.8 4.0 7.0 29.0$ 6.9 4.0 4.6 25.6$ 6.1 3.9 4.4 21.3$ 6.1 2.8 4.1» Wholesale Trade Retail Trade Finance, Real Estate & Insurance Service 4.3 14.4 4.4 14.2 4.6 15.3 4.7 16.1 4.7 16.9 3.4 10.9 3.4 11.9 4.0 14.9 4.2 17.4 4.8 20.5 0.6 1.9 3.0 8.9 0.5 1.8 3.0 9.5 0.4 1.7 3.8 10.8 0.4 1.5 4.1 11.5 0.3 1.8 4.5 11.9 100.0 100.0 100.0 100.0 100.0 Mining Federal Government State Government Local Government Total Source: Senate Fiscal Agency, Commission 1984 and Michigan Employment Security 68 Table 4.6 General Fund-General Purpose Year End Fund Balances (In thousands) Total Year End EAg.gaJ. Ysar__________________ Reyepug_________________ F_und._ Balance 1966-67 1,281,248 11,009 1967-68 1,644,964 55,903 1968-69 1,986,000 130,806 1969-70 2,114,681 44,745 1970-71 2,479,284 721 1971-72 3,013,083 31,338 1972-73 3,731,060 184,686 1973-74 3,964,879 207,186 1974-75 4,052,592 1,579 5,888,401 28,348 1976-77 5,383,347 67,888 1977-78 6,175,235 63,557 1978-79 6,419,529 267,169 1979-80 6,833,240 226,488 1980-81 7,153,776 160,128 1981-82 7,478,281 (635,982) 1975-76 — 15 mos. Sources: State of Michigan Financial Report, Dept, of Managment and Budget, 1967-1983. 69 executive recommendations for tax increases. Recovery brought rising revenues (realized through tese tax hikes) which supported the continued growth of state government. Between 1962 and 1972, GF-GP expenditures rose from $*176 million to $1,999 million. During the 1950’s, the annual rate of growth was 7.8 percent; during the 1960*s this had grown to 15.4 percent per year. The 1970-71 recession was the first which saw the implementation of Article V. Section 20 of the 1963 Constitution. This provision requires the Governor to submit a plan to reduce expenditures to the legislature whenever it appears that revenues will fall short of levels necessary to sustain spending at appropriated levels. Throughout the 1970’s a number of tax cuts were implemented which both reduced GF-GP revenue and placed limitations on alternative revenue enhancement strategies which otherwise would have been considered or even recommended in 1981. Tax limitations steps were taken in 1973 (Homestead Tax Credit and increase personal exemption), 1974 (local governments could extend from tax improvements to industrial plants), also in 1974 a referendum exempted food and drugs from sales taxes, and in 1978 (the Headlee amendment providing for limitations on total state revenue, a floor was specified for aid to local units, and reductions in millage rates if assessed valuations increased at a rate faster than the consumer price index for that year). These changes in the state's tax structure contributed to the "series" of revenue crises between 1972 and 1983. During these ten fiscal years, revenue fell short of appropri­ ation levels six times (1975, 1976, 1980, 1981, 1982, and 1983).6 In each case, necessary savings were generated either through accounting adjustments (borrowing from special purpose funds, erasing accumulated 70 surpluses, and even extending the fiscal year), and executive order reductions. It was not until 1982 that a tax increase package was approved. Even though the electorate seemed to indicate a reluctance or indifference towards tax limitations proposals included on election ballots, the policy of the Milliken Administration was to reduce the tax burden for Michigan citizens and businesses while reducing the scope and size of state both counts. government. ^ These recessions gave him opportunities on Brazer suggests that the assertion that most people were getting less for their tax dollars because of the shift of priorities to Social Services Corrections and Mental Health Programs is probably correct since single interest and minority groups tend to benefit most from services provided by these departments. He feels that it is especially ironic that a state government acting in such a conserv­ ative manner to the revenue crisis was given the lowest state bond rating in the nation. "Pleading poverty has, of course, long been a popularly accepted means of escaping unwanted obligations. Unfortu­ nately, a major consequence of [Michigan’s] approach has been to cast an unnecessarily and unwarranted harsh light on the state’s capacity to meet its obligations.1'® The reductions made through enactment of the seven executive orders are summarized in Appendix A. program cuts. Most of these reductions became permanent Many of these cuts were made during the allotment phase, well after the fiscal year began. Much flexibility was lost when reductions had to be absorbed in a six or four month time frame. Just as many officials were involved as reduction plans were drafted and approved, several approaches were used to develop these strategies. The 71 strategies themselves changed as the crisis worsened in 1981 and 1982. Patterns in these behaviors and in the way that various decision makers interacted have become apparent. What is interesting is that most of these reductions were made without direct policy or programmatic guidance from the Governor or DMB. Consequently, these plans reflect the attitudes and goals of middle level bureaucrats more than the desires or directions of the Chief Executive. A more profound impact on policy is possible through program elimination or consolidation than might be possible through new initiatives or strategies. If program eliminations are proposed and programmed in the agencies, government policy making may be traced to this level. If the central budget authorities failed to question agencies on these budget cut strategies, policy making would have been delegated to mid level career bureaucrats. Priorities and policies may have been revised at the agency level through a highly selective reduction process. This may be one of the more troublesome consequences of the budget crisis of 1979-1983. 72 Notes 1. House Fiscal Agency,Review of the State of Michigan FiscalBudgetarv Process. (Lansing, MI, 1976), p. 12-13. 2. Executive Office of the Governor, Executive Budget REcommendations. 1975-1976. (Lansing, MI, 1975), p. i. 3. Allen Schick, "A Death in the Bureaucracy," for a excellent review of the promise and failures of PPB systems in the Federal Government. 4. House Fiscal Agency, o p . cit.. p. 15. 5. Ibid.. p. 15. 6. Harvey Brazer, "Anatomy of a FiscalCrisis: The Michigan Case," Public Budgeting and Finance. Vol. 2 (1980), p. 136. 7. Ibid.. p. 138-139. 8. Ibid.. p. 140. CHAPTER V HOW REDUCTION DECISIONS WERE REACHED Introduction This chapter documents and describes how budget reduction decisions were made in Michigan between 1979 and 1983. developed from oral interviews This description has been with key budget participants in the 1979-83 reduction process. officials and In those instances where confidentiality was requested, comments have been combined in a generic sense to give a picture of how recommendations were proposed and agreed to, without giving specific department examples. While the decision sequence and calculus portrayed here may not describe how all budget analysts reviewed and prepared recommendations for Department reduction plans, these participants acted out their roles as they are described here in a class sense. It appeared after these interviews were completed, that budget analysts and department budget officers generally prepared recommendations in a manner expected by rational, incremental and process theorists alike. No single theory proved to be the most appropriate model for this decision process. In Michigan, the Governor is responsible for the integrity of the budget. If it appears that General Fund revenues will not sufficiently cover appropriations, an executive order is prepared specifying how and where reductions will be made to bring expenditures into line with 73 n anticipated revenue. Article V Section 20 of the Michigan constitutions requires the Governor to prepare these planned reductions for the legislature. The Governor plays a key role as the only government official responsible for such a plan and, just as in the case of the normal budget preparation process, sets the fiscal agenda. The Governor proposes and the legislature either approves or rejects; it cannot modify the content of an executive order reduction plan. The process through which reductions are proposed and implemented is as complicated and political as the annual budget process. A multitude of officials and analysts confer, interact, bargain before most of these reduction and conclusion decisions are reached. Decisions were proposed, analyzed, discussed and agreed to in three different organizational environments: branch, the nineteen departments of the executive Colleges and Dniversities, the Bureau of the Budget of the Department of Management and Budget (DMB), and the legislative fiscal agencies and appropriations committees. Just as in the case of the normal budget process, department revenue ceilings are established by the Governor; program funding levels are requested by the departments; reviewed and recommended by DMB; reviewed, modified by the legislature; and enacted into law and implemented by the Governor, through established policies and procedures. Executive Order Process The executive order reduction process is triggered by revenue if revenue estimates of the Office of Revenue and Tax analysis indicate that General Fund - General Purpose revenue will not be sufficient to cover appropriations. When this happened in 1979-80 few anticipated that it would mean anything worse than what happened in 1975-76. The 75 revenue forecasts for fiscal year 1981 expected a mild recovery from the recession of 1979-80. Instead, a steady and continuous economic decline lead to the enactment of five executive orders in a twelve month period. Within weeks of the submission of one set of executive order recommendations another notice was sent to the departments to prepare plans to cut their programs. All Budget officials concerned wanted to avoid unnecessary cuts if the expected economic upturn were to develop during the next financial quarter. The State Budget Director, as Director of the Department of Management and Budget, set the tone and established timetables and requirements for the reduction process which followed. In consultation with revenue experts and economists employed by the department, the general statewide reduction figure was decided on and analysts in the Bureau of the Budget were included in these discussions. Each was asked about a departments ability to absorb a reduction of some given amount before a target reduction level was decided on for each department (this process is outlined in Appendix B). Initially departments with lower priority programs (in the eyes of the Executive) were addled with the most serious cuts. The problem in 1979-80 was largely resolved using bookkeeping adjustments (250 million) and the Budget Stablization Fund balance that had accumulated in 1978 and 1979 fiscal years (over $260 million). that programs were not cut in 1979. This does not mean however Comparatively obscure programs (special studies, cooperative programs, and some cash transfers to other units of government and universities) were eliminated and the process of reducing the scope and ambitions of state government was begun. A conscious effort was made throughout this crisis was to change the fiscal philosophy evident from the late 1960's and early 1970's. 76 Before, analysts evaluated changing or emerging needs for state services, steps were recommended to assure that revenue sources were adequate to support these needs and demands. From 1979 on, fiscal policy makers attempted to reduce the extent of government to levels which could be sustained by General Fund revenue without tax increases or new revenue generation programs. There were several who advocated tax increases in 1979-80 but the mood and policy of the legislative and executive leaderships countermanded such proposals. If it were possible to do so, budget reductions would have been implemented over a much longer time span. The recession that began in 1979-80 at first was viewed as an opportunity to redefine the scope of State Government. However as the crisis worsened, time simply was not available to make these cuts in an orderly or systematic way. The second year of the crisis was generally resolved through accounting "adjustments" and school aid and higher education programs were reduced. In 1982, everything fell apart and significant pain was felt everywhere. The Governor’s office and the top management of DMB wanted to see that the cutbacks were distributed fairly and in an orderly way. This became impossible because instability of the fiscal environment and the speed at which the revenue fabric of the general fund began to disintegrate. Fair initially. share or incremental reduction techniques were used A general ban on out-of-state travel was Instituted and freezes on hiring were applied in all executive departments at the direction of the State Budget Director. Department directors were instructed to maximize general fund lapses while exploring alternative ways to support more service programs through new or increased user fees. Funding switches were proposed and some laws were changed to 77 enable the diversion of fees from bond funds to other operating programs which had been supported by the General Fund. were liberally interpreted. Accounting principles As related in the 1984-85 Budget Message, the Governor observed that the constitution prohibits deliberate actions which might cause spending to surpass revenues. "Weaknesses in state law, however, do allow certain bookkeeping adjustments that can make an actual state budget deficit seem to disappear. Until the state's fiscal year 1982 Financial Report there had been no formal acknowledgment of Michigan's true financial condition...the true magnitude of these deficits was not fully known, and until the cash crisis forced us to face the problem head on, the issue was avoided or ignored."^ The effects of these practices are illustrated in Figure 5.1 below. These deficits were largely concealed through increased borrowing to cover tax refunds and grants to K-12 and higher education. However, questionable these practices might have been, they enabled the reduction process to proceed at a nearly controllable rate. The description which follows is organized into two phases: the DMB phase, where the executive order is drafted and ultimately presented to the legislative appropriations committees and the Legislative Phase, where Executive Order reduction strategies are reviewed approved or rejected. DMB Phase Executive order cutbacks were made through the close cooperation of officials in DMB, the agencies and the legislature. As the crisis progressed, these relationships became closer and procedures minimized the inevitable personal conflicts that program cutbacks often generated. The DMB Phase of the reduction making process followed the general pattern outlined in Figure 5.2. Four primary groups were involved: the Figure 5. .1 STATE OF MICHIGAN GENERAL FUND BALANCES ( C o m p u t e d on the b asis of g e n e ra lly a c c e p t e d a c c o u n t in g p rin c ip le s ) TOO f; M I L L '74 0 - 100 1 - 200 75 '77 78 FISCAL YEAR 79 '80 '81 '82 f '83 '84 '85 - '86 u I ■ 300 i O ■ 400 N S - 500 - O ■ 700 - 600 - F ■ 800 - D - 900 - 0 ■1000 L L A -1100 - -1200 - R -1300 - S -1400 - Projected as of January, IS83 -1500 -1600 -1700 N O T E : Fiscal Years 1974-1980 are estim ates. Fiscal Years 1984-1986 are p ro jection s Source: "Budget M e s s a g e of the Governor, 1984-85 Fiscal Year," Department of Management ar ’ Budget, 1984.. Figure 5.2 Michigan's Budget Reduction Process Ftepartnient of Nianaganent and liudget Phase IMP - IM1 Budget Analyst 4. Reviews department budget requests from prior years. • History of program expansions • Funding diversification • Account flexibility • Fiscal detail 5. Informal contacts with the departents tray occur. 6. Reccmnends target reduction levels for each department - Stnte Budget Director 1. Determines Statewide reduction target level 2. Provide revenue estimates to House and Senate Fiscal Agencies and the Legislative Leadership. 3. Asks Budget analysts for their perceptions of an agency's ability to absorb a given target cut. 7. Suns revised ta rg e t le v e ls fo r all departments. If adequate to Department Director rrEet expected revenue sliortagc: S. Sends target reduction letters to department directors. 9. Instructs budget officer to prepare a reduction plan, and to estiunte the probable effects these cuts might have on the deportment. 10. Prepares response or appeal to the State Budget Director. ( Go to 6. ) Progrgn Manager Figure 5.2 ( Continued ) Michigan's Budget Reduction Process Department of Management and Budget Phase IM1 15. - Budget Analyst P H - State Budget Director Reviews department reduction plans and prepares Q B reccmnended executive order. 16 a. If acceptible: Incorporate into draft executive order. 16 b. If not acceptible: Return to Dcpnrtent director ( Repeat steps 9 - 14 ) 17. Executive Order draft presented to Governor. Sane adjustments may be made. 18. Submit Executive Order to the Legislative Quadrant and the House and Senate Fiscal Agencies. Department Director 13. Budget officer prepares department- response and follows priorities set by the department director. 14. Director sends department reduction plan to the State Budget Director. Program Manager 11. Prepare plan to lapse target • list inpacts • Prepare irrplementation plan. 12. Bureau chiefs review submissions from divisions and propose final reduction plan. 81 DMB Budget Director and Top Management, the revenue and budget analysts in the Office of Revenue and Tax Analysis (ORTA) and the Bureau of the Budget, the 19 department directors and budget officers, and the program managers of State Governments. The General Fund shortfall problem was aggrevated by rising welfare costs at the same time that the revenue picture was deteriorating. A budget gap developed six to eight weeks before any direct action could be taken. There was a general tendency to choose a middle ground and not to forecast the worst case. During FY 1982 when there were four executive orders, there was an inclination to underestimate the degree of the problem. This caused a succession of executive orders totalling almost 800 million dollars and also included a temporary tax increase. Both DMB and the fiscal agencies were aware of these problems as they developed, and attempted to measure it with reasonable accuracy. A plan was then devised to close the gap. "We simply did not know how bad it was going to get, how bad it could get, and looking back at the situation things got so much worse than we had thought was even plausible. It is the only period in my memory where people doing the revenue estimates were unable to predict the direction of the change in even a preliminary way...and that hasn't happened to my knowledge in 30-40 years in this state." (DMB Deputy Director) Once the general nature of the problem was determined by ORTA, and the state wide reduction figure was agreed upon, the budget analysts in the office of the budget assessed the ability of each department absorb its share of the cut. to Each analyst was to propose a target reduction that could be made without seriously effecting the general missions of priority program areas. In most cases these exempted areas were given high priority during earlier executive budget cycles. The analysts would then recommend a reduction target for each agency to the 82 State Budget Director and these then would be sent to the Department Directors. The department target reductions were based at least in part on Management 1979-1980. Plans prepared for fiscal year 1977, 1978, 1979 and In each of these years, departments were required to submit requests sensitive to three different levels of General Fund support (generally 90, appropriations). 100 and 110 percent of adjusted current year As one can see in the sample request summary for the Forest Management Program (Table 5.1), it would be possible, with several budget requests to estimate the amount of flexibility in the department's budget. Since the departments were to describe the effects each funding level would have on program impacts and on its ability to carry out its responsibilities, a DMB analyst could estimate possible effects of a ten percent program cut. Reduction targets derived through an incremental calculation e.g. based allocation. on 5 percent DMB would of a department's gross general fund first look at the amount of flexibility a department might have and how important their operations were in the context of other departments. The departments would be given an opportunity in advance to say how they might come up with their possible reduction target. While it appeared from some perspectives to be a rather crude process, department targets and the executive orders were prepared in a very selective way. Discussions were sensitive to the Governor's priorities, the effects these cuts might have on the public, and effects these cuts might have an agency's ability to provide services. Informal contacts were made between budget analysts and department budget officers, and fiscal agency staff. The fiscal agencies were 83 Table 5.1 Department of Natural Resources Forestry Appropriation Unit Detail Base Current Year Appropriation 90$ 5,727.2 100$ 5,727.2 108$ 5,727.2 3.9 54.2 54.2 Department Requested Changes 1. Step, Reallocation Adjust. 2. CSS&M Adjust. 222.5 (23.2) 1.9 8.9 3. Travel Adjust. 185.0 (16.3) .8 5.8 4. Equipment Adjust. 132.5 (82.5) (62.5) (32.5) 5. Special Maint. Adjust. 15.0 (10.0) (7.5) (5.0) 9.7 9.7 9.7 - - 6. CFA Payments 7. 313.6 Forest Cultivation Adjust. (113.2) 8. Federal Pest Project 9. 5.2 Tech. Asst. Program 10. Timber Sales PRR #3 5.2 5.2 (35.5) (25.5) 8.5 246.4 246.4 246.4 11. Regional Economic Development PRR #32 50.0 12. Forestry Mgt. Fund Ad. PRR Total GF Source: $ 5,727.2 2,822.6 5,711.7 3,070.5 5,949.9 3,308.7 6,078.4 3,437.2 Budget Worksheets, Department of Natural Resources, 1978. 84 informed of changes in revenue forecasts which required an executive order, but the target aspect of the process (the designation of agency reduction levels) was a DMB effort. Departments were given maximum flexibility to come up with the dollars. The line item detail was not asked for until the Executive Order was approved and implementation was begun. While it is reasonable to expect a department to develop a reduction strategy out of the detail, this information was not required by DMB until much later in the process. The way that these reduction orders were reviewed and responded to varied from agency to agency. The management style of the director, the amount of autonomy granted to divisions and other units of the depart­ ment in fiscal matters and the characteristics of the agency budget process in place prior to the fiscal crisis each influenced the way that an agency’s plan was prepared. Departments with a highly centralized system of program control (e.g. DSS, DMH and State Police) the depart­ ment top management and the budget officer would prepare the departments response. Others with a more distributed responsibility system tended to involve the operating divisions and field management units in these decision. In centralized departments, reductions were more in the form of deferrals than program cuts. replacement schedules By this it is meant that equipment (including motor vehicle replacement) were suspended, capital outlay projects were canceled or delayed and contract purchases of services picture improved. general growth or products were cancelled until the budget In those cases where the budget had been experiencing (in real terms, not just inflationary increases), reductions mandated through DMB were taken in approved expansions. This 85 made it easier for these high priority departments to cope with cuts made in 1980 and 1981. By the time FY 1982 came around those methods proved to be inadequate and cuts in base programs had to be made. Departments which were particularly high in the Governor's priorities, like the Departments of Mental Health, Social Services and Corrections, were offered opportunities to spread these cuts over several years. This tended to aggravate the state's cash problem but it enabled these departments to provide sustained and predictable service levels for those in need. Some of these reduction plans were developed jointly between DMB and the department. For example one department was encouraged to use anticipated federal money to replace General Fund appropriations through an accounting procedure called an expenditure credit. Normally, the allotment of federal funds is handed through a supplemental appropriation. Unfortunately, the details of this agreement wee not communicated to DMB or department accounting staff and as a result an over-expenditure of several million dollars occurred. The General Fund target was not allotted, resulting in the overage. The technique of deferrals, while painful and dangerous in some cases, generally worked well for these agencies and it has been relatively easy to have these funds restored in the 1984 and 1985 fiscal year budgets. Reductions in the larger departments were developed centrally. Directors required such steps as the reduction of overtime by one half, appointments would only be made if approved by the Director. Actual dates of hire would be extended as far as possible and equipment purchases were suspended. Director. Exceptions could only be made by the Through the benefit of hindsight, some budget officers feel that they might have been more successful if a management plan approach (evident in the Governor's Executive Budget process) had been used, both 86 internally and as a way of presenting their reduction plans to the Governor. Effects of Structure on Strategy As the crisis worsened, the strategies of centralized departments began to change. This happened for basically two reasons. First, the flexibility offered through approved expansions and funding switches had been exhausted during the first several executive orders. These depart­ ment directors needed the assurance that the cuts would be implemented in such a way that the organization would still be able to function and address priority service needs. With these objectives and realities in mind, these departments turned to a more programmatic approach to budget cuts. In some instances clients were diverted from the state system to local care givers — local government and licensed care agencies. These termination decisions were made in a fairly thoughtful and objective way. Generally, there were not major breakthroughs in the efficiency of the programs that remained. Some functions were merged as bureaus were reorganized, but the organizational response to cutbacks tended to be more bureaucratic and political than economic and "rational". The approach most favored by these budget officials two years after these reductions took place parallels closely with that used by de-centralized and smaller departments. Departments with distributed accountability in their standard budget process and accounting systems both desired to (and out of necessity had to) decisions. involve division and program managers in cutback Generally, the process unfolded this way. Once the depart­ ment reduction target was received from DMB, a reduction plan was required of all divisions partially or wholly supported from General 87 Fund revenue. If the department's target was one million dollars, a table of division targets adding up to 1.-U million would be prepared. Each division would have to respond to its assigned figure. Divisions in decentralized organizations enjoyed a great deal of flexibility especially if no strategy constraints were placed on it by top management. Two financial plans were used by these divisions - the management plan and the allotment or spending plan for the current year. The management plan addressed priorities and program impacts of varying levels of financial support. Allotments reflected the actual rates of spending and special purchases by program managers. During the early years of the crisis, division chiefs reviewed existing vacancies, equipment purchases, operational costs of field and staff units, travel patterns and alternative funding sources. Special interest projects, including those placed in the division's budget by the legislature (with or without the division request to do so), were included. All of these items were then evaluated for possible lapse savings given the time remaining in the fiscal year and the potential effects on the division if these cuts became permanent. All feasible alternatives were then ranked in priority order, and those low priority items were skimmed off in levels of increasing priority until the required cut was reached. During the first several executive orders, this technique tended to concentrate on a program's financial needs and on the means to be used to generate needed savings or lapses at year end. The focus changed from the inputs (staffing levels, logistical needs and hiring plans) to program outputs and public benefits as savings become increasingly more difficult to find. In nearly all cases (while most division budget officials are unable to say precisely when), 88 a point was reached when discussions changed from "how might programs be streamlined further through fiscal and procedural changes to," what programs can this division afford to support and sustain." This change was a significant and important one in that it required a closer look at program beneficiaries and the priorities. The division reduction plans reflected a program focus and this was reflected in the departments' responses to DMB. In some departments, the process described above was repeated as each division plan was submitted to bureau chiefs and to the department budget office. Department priority lists had previously been developed through the management plan process and were used to prepare the Department's reduction plan. Some departments asked for division cuts which when added together, provided amounts higher than the DMB target. This enabled department top management to select those reductions which would generate the least resistance and had the smallest impact on department missions and goals while protecting other areas. Some division recommendations were rejected for this reason, but most were endorsed and embodied in the departments strategy. The detail and analysis produced in this process is most evident at the division level of the organization. Department reduction were prepared through a bottom-up process with little guidance provided by higher bureaucratic levels. Exceptions were made on a case by case basis. Reduction strategies at the lower levels in the hierarchy resulted in considerable variations in allocations within appropriated program levels. In some cases supplemental requests were submitted for restricted fund programs after general fund cuts were made. As these supplemental were approved, position vacancies became available into which displaced employees were transferred (or bumped). In terms of the gross, no overall funding level change occurred. A funding switch was the real consequence of what was thought of before as a program reduction. In other instances games were played with funding sources. Federal funds were masked as general funds and included in program cuts. As the grants were awarded or received, supplemental requests restored these "federal programs". For one instance, General Fund levels were cut by $20 million but in reality only $10 million was actually saved. Program Impact Information The usefulness of the department management plan in this process depended on how seriously the department approached it at the time. In the opinion of the fiscal agencies and DMB, the usefulness of program performance information was very uneven. This might be traced to the perceived importance of this information during normal budge cycles. In those cases where it made a difference or dramatized the importance of a program, it tended to be included in reduction plans. However if the measures failed to show a connection between program levels and public benefits, it was generally lacking in impact statements. The narrative appears to be the greatest casualty in the executive order process. If one agency can be used to generalize the process, three and four page impact summaries were digested into two or three paragraphs in the department package. When the completed Executive Orders were finally submitted to the legislature, none of the fiscal detail or the narrative was initially available to the appropriations committees. If program narratives were generalized too much, the specific program impact of each itemized reduction was not available. 90 Those agency reduction plans approved by DMB were included in the draft executive order. If a department's plan was unacceptable it was sent back to be revised. The cycle could be repeated until the final draft executive order could be approved by the Governor. tuning occurred Some fine before an order was sent to the appropriations committees^ Legislative Phase There often was a high level of interaction between DMB and the legislative leaders to assure that the order would be approved as DMB had it prepared, but this was not always the case. There were times when orders were simply "dropped" on the appropriations committees without any preliminary discussions. These were disapproved and then a negotiated order was worked out afterwards. The process varied from time to time, but there was more cooperation with legislative leadership as the crisis worsened. Many gimmicks were used to reduce as much as possible the harm that might have occurred if the dollars had been taken away at one time (e.g. deferred payments to manage over a couple of years what otherwise would have had to have been handles over four to six months). This was very important since many of these orders were issued in spring or early summer. The state's fiscal year begins on October 1 and runs through September 30. The appropriations committees had ten days to either accept or reject the order as presented — they could not change it. If the legislature wanted to change an order, it would first reject it and then negotiate a revised executive order. Some negotiation discussions took place in advance of formal presentation but in all cases the legislature eventually approved an order proposed by the Governor. Program performance information was not generally available to the legislature, but information was not generally available to DMB either. The state of the art had not progressed much further than where it was in 1977-78 when the target budget approach was abandoned in favor of a priority budgeting system (modeled zero-based budgeting). after other state's experiments with The absence of information of any kind was lamented by fiscal agency staff. The information that was available was very sketchy and really was not used to review executive orders. The fiscal agencies did not have the impression that these reduction targets were arrived at through any systematic review of program efficiency or effectiveness. They saw no priorities in these DMB submissions but a mix of rational budgeting and raw politics. were evident but the The Governor's priorities final reduction plans also reflected the distribution of political influence in the executive and legislative branches (comments of fiscal agency staff). As a draft executive order was received by the appropriations committees it was split up into component parts corresponding to depart­ mental assignments among the appropriations subcommittee chairmen. Each chairman reviewed the order's effect on departments programs slated for reductions with fiscal agency analysts. subcommittee would report After this review, each its recommendations and reservations, if any, to the appropriations committee leadership. Alternatives to the reduction order were considered and noted in each report. These analyses were then presented for the consideration of the "Quadrant''. The Quadrant was the principal policy making body of the legislature on fiscal matters. This group consisted of the Speaker of the House, the Majority Leader of the Senate, and the minority leader of each house. These officials were supported by the two appropriations 92 *• committee chair, the majority and minority vice chair, and the Finance and Taxation Committee chairmen of each house. caucuses were also represented. The urban and black Quadrant meetings would also be attended by representatives from the Executive Office and DMB. The representation in the Quadrant membership was very important. If this group could agree on the executive order, the balance of the legislature could be expected to go along with its recommendations. "This group would meet, and conflicting revenue estimates might be presented by the fiscal agency directors. around and come up with an agreement.n The group would kick it This agreement represented the reduction order. Negotiations became more intensive in the later stages of the recession (in 1982 especially). The orders became bigger and they came on top of cuts that had already been made. Even though some gimmicks were used to take maximum budget advantage, the negative effects could not be avoided. As it became increasingly painful both programmatically and politically, there was an attempt to draw more legislative actors into the process in advance. This gave various interests at least an opportunity to make their views known. Some factions in the legislature sought to change the program areas effected by executive orders — — and advocated tax increases. cut education less, social services more From the perspective of DMB there never was a coalescing of legislative majority support around a proposal that was very different from what the Governor had proposed. the process played an important part. The politics of Squeaky wheels were greased. External constituencies (e.g. local school districts and universities) had an effect on the amounts and the timing of some items in these executive orders. 93 The media and client activists were unable to influence these bargaining sessions. Most of the details had been hammered out before either of these groups sought to change the outcomes. "There had been a lot of private discussions both entirely within the Executive Branch and with the involvement of the Legislative Branch. Their concerns were fairly well known and in some cases orders were approved as if the rubber stamp was ready before we entered the room. Everybody knew what was in the order, nobody liked it and they wanted to get done with the business as quickly as they could. The media did descend in droves on these presentations and we had some fairly lively discussions in a couple of cases. But for the most part it was known before we walked in the door that it was going to get approved. The legislators simply did not want to expose themselves to the unpleasantness at hand." (DMB budget official) Constitutional Provisions Section 20 of Article V of the 1963 convention (pertaining to fiscal management responsibilities) Michigan Constitution. was not included in the 1908 It specifically addresses situations which might require immediate action to keep expenditures in line with anticipated revenue. Sec. 20. No appropriation shall be a mandate to spend. The governor, with the approval of the appropriating committees of the house and the senate, shall reduce expenditures authorized by appropriations whenever it appears that actual revenues for a fiscal period will fall below the revenue estimates of which appropriations for that period were based. Reductions in expenditures shall be made in accordance with procedures prescribed by law. The governor may not reduce expenditures of the legislative and judicial branches or from funds constitutionally dedicated for specific purposes. The first sentence of section 20 provides the Governor with an opportunity to take advantage of efficiencies or economies which might not have been expected when appropriations were initially enacted. second sentence requires the Governor, The with the approval of the legislature, to reduce expenditures whenever it appears that actual revenue collections will not be as much as preliminary estimates stated. It was the intent of the constitutional convention that this sentence would remove any question as to the constitutionality of legislative control over the state’s fiscal policy. It requires immediate action to minimize impeding year-end deficits. The final sentence protects the separation of powers doctrine by preventing executive reduction of expenditures for the "coordinate legislative and judicial branches of government."^ It also prohibits the Governor from reducing funds dedicated to specific purposes by the Constitution. In comments on the majority report of the proposed draft of this section of the constitution, Mr. Austin said, in part, "I think more than any other issue it was the problem of revenue shortage in the state of Michigan that prompted the people of the state of Michigan to order this constitutional convention. The people were somewhat concerned that the legislature either did not have the power or refused to exercise the power to raise sufficient revenues to pay the cost of government... Now what we are seeking to do is to provide an escape hatch for the legislature by putting the ones, ad Delegate Marshall has indicated, on the governor to reduce expenditures where the legislature has failed to provide sufficient tax revenue."3 Speaking for the proposed change, Mr. Martin (a Republican from Grand Rapids and Chairman of the Working Committee) emphatically expressed that it was "intended that the legislature could exercise a supervisory responsibility over any reductions that the Governor might make and just specify how he should make such reductions."^ Proponents of the minority report (suggesting among other things that section d not be adopted) felt that it was not good government policy to require state budgets to be balanced on an annual basis. should be a surplus and years where "There are years when there there may be a deficit." 95 (Mr. Downs, p. 1678 Con Con Record). This opinion was not sustained by the convention and section d was included in the final draft. It appears that it was the intent of the convention that the legislature and the governor should be jointly responsible for keeping expenditures in line with anticipated revenue. The earlier constitutional language stated that the legislature was responsible for raising revenues in sufficient quantity to support mandated expenditures embodied in the appropriation acts. This shift from management of tax revenue to control of government expenditures is clear. From this discussion it is apparent that Governor Milliken was justified and was on firm constitutional ground when it was decided that government expenditures would be reduced in lieu of a general fund tax increase. This might also explain the current policy of enacting a tax increase package for a term sufficient to generate enough funds to erase the accumulated deficit. The current policy of preparing budget recommen­ dations within expected annual levels of revenue is very similar to that followed throughout the 1970-1983 period. Restricted funds like the General Fund are sustained by revenue sources which do not always generate enough to support appropriations. It appears as a result of this research that section d provisions are not followed in cases involving these special funds. Instead, depart­ ment directors are responsible for maintaining a balanced budget for programs dependent on these sources. Procedurally, this is like other fund maintenance practices as they effect non-state revenue sources lapse amounting sufficient to cover deduct when collections fail to match anticipated levels. While it is not being suggested that this practice is somehow unconstitutional, it shows another way government approaches revenue problems. Since there is no legislative review of these reduction plans, there is no way of assuring that cuts will be made in a manner consistent with the wishes of the legislature. writing, As of the date of this the executive order process has not been used to resolve revenue shortages for funds other than General Fund-General Purpose. Management of these other fund balances appears to be an exclusive executive branch matter. In most instances, department lines. Apparently this is the case for two reasons. these funds support programs that do not cross They may have been created to provide user-fee support for programs (e.g. state parks - Park Fee and Concession Fund, harbors, docks and access sites - Waterways Fund) or to provide funds to support regulatory programs. Since administrators can control expenditures in these areas, no one in the legislature or DMB is in a big hurry to include these in their revenue projections or expenditure control programs. Secondly, these funds represent a very small proportion of the state’s operating budget. It would be fairly easy for fiscal analysts to become emersed in these specialized areas. If this happened, obviously higher priority assignments could be effected. It is both logical and easier to leave special fund cash management to the departments and to concentrate on General Fund programs. 97 Notes 1. Executive Office of the Governor, 1984-85 Budget Message (Lansing, MI, 1984), p. 5. 2. Constitutional Convention, Letter to the People. (Lansing, MI, 1963), p. 49. 3. Constitutional Convention, Record. (Lansing, MI, 1963), p. 1658. Section 2 of Article X of the 1908 Constitution placed this responsibility on the legislature. It was to provide sufficient revenue to support the estimated expenses of government. Section d of the draft proposal addressed by Mr. Austin shifts this responsi­ bility to the Governor. 4. Ifcid., P. 1959. CHAPTER VI SUMMARY AND CONCLUSION Introduction No single theory exists which describes how budgets are reduced by state governments. The budget literature suggests that the reduction process might follow the general decision pattern outlined in the rational/comprehensive, the incremental, or the process models. It is the hypothesis of this study that targets for budget cuts are proposed incrementally, analyzed rationally and comprehensively, proposed and discussed through bargaining process, and summarized and acted upon incrementally by the legislature. Based on the recollections and responses of key budget officials, reduction decisions were not consistently made as the process and incremental models suggest. Some departments used a priority approach throughout the process while others tended to use a target approach based on a program's proportional share of the departments total general fund appropriations. progressed. Some agencies changed strategies as the crisis The program cuts offered by the departments established the agendas for subsequent discussions. Only marginal changes in these plans appear to have been made along the way by DMB and the legislature. Interview Results 98 99 The results of the oral and written responses to the interview questions are summarized in Table 6.1. significant agreement in all four categories. It appears that there is Most reductions were: 1. arrived at through a target approach based on share of the total agency budget on a programmatic basis. In nearly all cases, reduction strategy changes were made as time progressed. 2. discussed in organizational and program impact terms. Program performance information was valued by all but fiscal analysts in particular lamented that this information was not always very useful. Program priorities of agency directors often were endorsed by DMB analysts and respected by fiscal agency analysts. Planning issues (future benefits or effects on anticipated demands for services) were clearly secondary concerns. 3. influenced by those officials within their range of authority. While their recommendations were important in the other two phases, the absolute dominance of these officials in their specific areas was consistent with their leadership roles. Table 6.2 is much like Table 2.2 in the except that those aspects evident budget reduction processhas been highlighted. It should be noted that the reduction process evident in Michigan did not fit any single budget model. The Process Model seems to best describe how the process worked, but there are some important departures from process model propositions. The rational/comprehensive model still describes the analytical basis for many of these cuts. The incremental model provides an accurate picture of how the detail of the reduction plans was developed. Executive order cuts were the result of hundreds of decisions produced through a politicized bargaining process. these decisions were produced through a satisfying process. orders evolved out of cautious incremental steps. the range of possible cuts and selected Many of Reduction Officials reviewed the first acceptable 100 Table 6.1 Interview Results Proportion af Agreement Ditension Agency Officials n=7 Fiscal Analysts n=13 State Budget Officials n=7 Concensus n=27 Strategy: Reduce by Prograt ' Reduce by Share (Target) Strategies Changed with Tiae 1.00 1.00 0.71 0.92 0.85 1.00 1.00 1.00 0.43 0.96 0.93 0.78 Decision Units: Organization Structure Proqrat Structure Appropriation Unit Structure Iteas of Expense 0.71 1.00 1.00 0.71 0.92 0.46 0.85 0.38 1.00 1.00 1.00 1.00 0.89 0.74 0.93 0.63 Relevant Intonation: Appropriation History Fiscal Detail Departtent Priorities Prograt Perfortance Issues and Opportunities 0.14 1.00 1.00 1.00 0.57 0.85 1.00 1.00 1.00 0.23 0.29 0.43 0.71 1.00 0.57 0.48 0.85 0.93 1.00 0.41 Ability to Change or Enforce: Agency Officials: Agency Phase DMB Phase Legis. Phase 1.00 0.71 0.00 1.00 1.00 1.00 1.00 0.43 0.57 1.00 0.78 0.63 Agency Prograt Manager: Agency Phase DMB Phase Legis. Phase 1.00 0.29 0.43 1.00 0.00 0.54 1.00 0.43 0.29 1.00 0.19 0.44 DtlB and Legis. Fiscal Analyst: Agency Phase DMB Phase Legis. Phase 0.43 0.43 0.29 0.15 0.77 0.46 0.43 0.86 0.18 0.30 0.70 0.41 State Budget Officials: Agency Phase DMB Phase Legis. Phase 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 101 Table 6.2 Michigan's Reduction Process Rational/ Comprehensive Reduction Issue Strategy .-Modal_______ Decision Units Trogr3TT!Temen?^" Reduce By Program vs Funding Levels Issues Development Long Term Strategies Short Term Issues Eudget Official Influence Priorities Used Dominant Decision Rule Program Efficiency & Effectiveness Process Model Progressive Cuts Program Cuts Incremental Current Tear Funding Level Current Tear + Growth History IncremeSaT^^ Incremental Adjustments Decisions From Stage to Stage 7ric?rT?Iea of Dominant Official Given Stage of Process ?air Share & Progressive Cuts By Program No Priorities Only Influence I Situations, doles,j Elected Officials I & Decision Making . Decide, Others | Environments ' Recommend i Dominate I — — .J Dominance Assignment • chief Executive I TImportant EssrarsComponent rwssr" Different Officials|TegislaTors Depending on Stage | Opportunity Costs May Be Taken Into Account Opportunity Costs Rarely Considered I 102 alternative, in most cases. The Michigan reduction process followed the pattern described below. Strategy The general focus of these decisions rose from the line item detail to program issues as time progressed. Early decisions were targeted, that is they were influenced by a program's share of the total agency budget. Decisions tended to affect program areas a general in an across-the-board fashion. As reduction decisions became more difficult, subprograms, programs and finally program areas were targeted for general cuts. With this progression came a greater concern for both agency and statewide priorities. Officials wanted to be confident that those programs remaining after the crisis passed would be those which provided the greatest public benefit. Some programs were reduced progressively, that is eliminated in stages. The easiest cuts were made in 1981 and 1982. By the end of the 1982-83 fiscal year most pork barrel and out-dated programs had been eliminated by the departments themselves. Each time a reduction order was received, a larger cut was taken until the minimum operating level was reached. Quite often, this minimum was lower than that initially suggested by the agencies. Decision Units Budget cuts were made on a program basis. The basic analytical unit was the program element (that is the lowest level where discernable services can be attributed to a single program manager). Each fit conceptually into the organizational structure of the department and 103 into the appropriation structure of the operating budget. Program line items are slowly replacing appropriation units based on items of expense to facilitate program decisions. Programmatic reductions were not a consequence of the severity of the cuts themselves but were predominant in later stages because other strategies were no longer able to produce necessary savings. Program choices are preferred over organizational capacity issues (outputs vs. inputs) by budget officials. The availability of program performance information is not 'the critical determinant of the program choice mode of decision making. Issues Development Decisions were made incrementally with adjustments and compromises made at each stage in the process. They were both directed to short term and organizational issues and to effects on priority programs. The macro focus described by LeLoup is was in much evidence during this reduction period. These impacts were reviewed in a comprehensive manner as the leadership became involved and a general planning or policy orientation supplanted the micro and short term focus evident during the early stages. Budget decisions were centralized by three key groups: Directors, Quadrant. Department the State Budget Director and DMB, and the Legislative Supported by budget officers, fiscal and revenue analysts, principal deputies and the appropriation subcommittee chairpersons, reduction decisions evolved out a bargaining process and were ratified in the final analysis by the legislature. This encouraged a "priority" 104 focus for decisions sensitive to the comprehensive policy preferences of the Governor and the Legislature. Many reduction decisions were the result of a rational process of choice based on missions, objectives and priorities. The priorities of dominant officials were reduction plans were consolidated. evident as in the Institutional goals and priorities were especially important as department directors prepared their budget cut proposals. director, As a reduction was legitimized by the DMB and the "Quadrant", these preferences tended department to be more personalized. The organizational climate, the timing of discussions, and the roles of various officials influenced reduction decisions. however, In general those officials represented at the final stages of the executive order draft made the hard choices. key role in these matters, decisions. While bureaucrats played a all events lead to final legislative The elected officials ultimately decided where the cuts would be made. Others prepared plans for how these cuts would be made. Bgainanss Each stage of the process was dominated by the preferences of a principal decision maker (see Figure 4.1). The program cuts proposed in these draft reduction orders were prepared and ultimately implemented within the departments by program managers. In so far as only marginal changes were made by DMB and the Legislature, it might be suggested that •reduction plans were a product of an agency process. No cuts could be 105 made however without the consent of the "Quadrant" and the Governor. Assignment Decisions in FY 1980 and 1981 attended to be focused on department or state-wide spending practices: travel was restricted, equipment was deferred, hiring freezes were mandated and accounting practices were liberally applied. Attention was focussed on items of expense, for example salary and fringe benefits, supplies, contractual services, travel, and special projects which crossed program lines (in those cases when several divisions shared a common line item). departments presented programmatic reduction plans. Later on, more Since these cuts were presented to the leadership on a programmatic basis, discussions and decisions focused on program issues instead of on the fiscal detail. Reduction alternatives were often discussed in terms of opportunity costs and impacts were discussed in general terms, but reliable information on the size, extent and ultimate consequences of these cuts was largely absent from the process. Decision makers would have liked to have had this information, but it was either beyond the capabilities of the bureaucracy or impossible given the time available for any kind of analysis. Analysis is always expensive, but the time costs proved to be the greatest deterrent to systematic program analysis. All budget officials expressed a desire for solid program impact information related to priorities, objectives and services. Reduction decisions were reviewed by analysts and subcommittee chair who had the best understanding of these programs took over. Those analysts who knew their program areas well had the greatest influence over reduction outcomes. 106 Most budget decision packages were prepared at the lower levels of the agency hierarchy. The matter of influence over outcomes could not be resolved from data generated through this study, but there is reason to believe that departments (i.e. program managers) are the key element in understanding how budget choices are articulated and program plans are prepared. Very little detail ever surfaced during as these executive orders were drafted. When minor adjustments were asked for (as they were in 1979 and 1980), assistant decisions were based on the fiscal detail - staffing levels and support requirements. Major adjustments were decided as bundles of smaller decisions on a program basis. The reduction process encouraged agencies to prioritize their programs. The Executive Order process of proposing and endorsing General Fund cuts worked fairly well. Cuts were not distributed across the board in a blind or mechanistic way, but were instead directed toward lower priority program areas. Pork barrel projects did not survive at the expense of other services. In general, officials feel that those programs which have not yet been restored (1985) will probably not be reinstated in their original form. The impact of these cuts on State Government was severe, but matters could have been much worse if they had been instituted over a two year period. No signs emerged which might indicate that the budget system and program priorities of the mid-1970’s were anything but what the leadership wanted. Reduction decisions were influenced by political, institutional and economic variables. 107 Political Variables Several officials described the political aspects of reduction decisions. Most of them described methods they used to influence the final outcomes. personnel. Information was subsidized and filtered by department There also were many opportunities to take advantage of the slack in the system. The timing of submissions proved to be critical since some decisions were made at only one point in the process. Decisions were not reviewed comprehensively at each stage, and efforts to influence deliberations were evident throughout the process. The most successful brokers in these decisions knew who to speak to and what information was crucial for their decisions. They also seemed to know when their input could be the most effective. The most important independent variables in the decision process could be the strategies selected (1) by the agency director to assign reductions and (2) by various participants as they attempted to modify these cuts or to divert reductions to other programs areas. Political success might be measured in terms of the number of agency programs protected at the expense of other department programs and activities. Impact Information Budget officials are not technical specialists. They usually deal with broad program areas and are involved in a diverse collection of issues at any one time. program area. They may be infrequently involved in a given As a result, they often are limited by inadequate under­ standing of a programs service delivery process. This can result in unclear explanations of possible effects of proposed program cuts. There was a general reliance on information brought forward by others. 108 Those who supplied this impact data had the greatest potential influence. Agency Prosperity There was no back lash directed toward those programs which had experienced sizeable growth between 1974 and 1979. DMB officials and fiscal analysts encouraged some departments to use recently expanded areas to absorb reduction targets. They could have been taken arbitrarily before departments could have responded, forcing them to make real cuts in previously established areas. Distribution of Cuts Eighty-five percent of the General Fund budget is appropriated to six departments, higher education and K-12 school aid. While priorities are evident in both draft and final executive orders, the remaining fifteen percent of government did not have to pay for the prosperity of the others. If no cuts were made in the 85 percent group and the reductions were made in a single 12 month period, all of the remaining thirteen departments would have been permanently shut down. Instead, cuts were spread over time and a rational explanation for year-end deficits was provided. A general spirit of realism and resignation prevailed over DMB and Quadrant reduction hearings. Political variables evident in other budget process studies have not shown an effect on reduction outcomes: size, structure, budgetary constituencies, agency age, history or gubernatorial support. Of these executive support seemed to be more important, but reductions reflected a general consensus among all parties more than any single political or 109 personal Ideology. This might be attributed to the manner that reductions were presented and discussed. no program area was untouched. The crisis was so severe that It was not as much a question of which ideology might prevail as who can be protected from the effects of these cuts. Institutional Variables Agency Size Those departments which were large, complex and highly centralized seemed to have the most trouble making agreed upon cuts stick. This reflected a general difficulty in controlling spending because of problems with internal communications. In other instances, conditions changed so quickly that actions once considered possible could no longer be taken without jeopardizing critical services to those in need. Program Complexity While not documented, there is reason to believe that departments which had widely dissimilar programs and service groups tended to use priorities, goals and objectives to prepare reduction plans. Those areas where programs were more alike were cut without concern for their relative priority. gtidKgi-Sfcr.UQtmr.g In those cases where budget bills contained program line items (vs. tradition line items of expense appropriation units), reductions tended to be discussed in program terms. How a budget was structured influenced how budget issues were discussed in DMB and the legislature. Related to this is the degree that a department might have a diversified 110 source of financial support from restricted and special purpose funds. Funding switches were made to reduce a program’s dependence on the General Fund. Program expansions enabled agencies to shift employees from eliminated programs into other areas. This made it possible to protect expertise and the organizational integrity of the department. Economic Variables Changing Revenue Forecasts - conditions were changing so quickly and unpredictably in 1981-82 that revenue analysts were not able to predict the direction of change let alone its magnitude. It did not make sense to enact a budget when no one was able to give even an approximate picture of the revenue situation. This indirectly resulted in the delay of the beginning of the allotment phase of the 1982-83 and the large number of executive orders enacted during the spring and early summer of 1982. Revenue uncertainty nearly created a totally unmanageable situation. Propositions Conclusions of this study may apply to other states when the range of variables evident in Michigan between 1979 and 1983 interact in similar ways. This study may also be inconclusive. why the outcomes must have developed as they did. It could not show It only can offer why it is likely that the cuts were made as they were. The explanations themselves need to be explored to determine if they are limited to this particular context. Based on the interviews conducted with officials in the legislature, DMB and the departments of the Executive Branch, the model 111 which appears to fit the reduction process in Michigan is shown in Figure 6.1 Figure 6.1 — Relationships and Outcomes of the Executive Order Reduction Process Department D irectors Governor Other Interest Groups Legislature Department o f Management and Budget Reduction Order In addition to the primary participants, other interest groups (at times with the encouragement of department directors and at others acting alone) attempted to influence reduction decisions. officials successful. interviewed felt however that these groups None of the were very According to legislative sources, organized demonstrations on the Capitol building steps tended to solidify positions more than change proposed reduction targets. Few positions or votes were changed. Figure 6.2 illustrates the relative influence each of these groups seemed to have on the final outcomes. The legislature Quadrant and the State Budget Director were by far the most important participants in the process. 112 Figure 6.2 — Group Influence over Reduction Order Outcomes Departments Governor L e g is la tu r e Other I n t e r e s t Groups Reduction Order »•••*; As described earlier, the officials responsible for the outcome of their phase of the process had the greatest influence at that time. Since no reduction order could take effect without the approval of the legislature, the legislative leadership through its veto power had the greatest impact on the reduction order. If one looks at the content of these orders and the detail of the executive order cuts, the Governor’s budget staff set the stage and placed their power and prestige behind his plan. There was no question in any one's mind that the Governor set Michigan's fiscal policy and only rarely did serious disagreements on the technical aspects of the policy develop. The architect of the plan was the State Eudget Director acting as the Governor's chief economic advisor. The department director had the greatest control over the effects of these cuts on the agencies. 113 Most issues were related to the allocation of the reduction targets among departments agencies. instead of the specific programs reduced by the There is reason to believe that department directors do have a disproportionate influence over which programs are protected, reduced or eliminated altogether. During the executive order process program elimination recommendations attracted more attention than others. It should not be surprising that program elimination recommendations are questioned more carefully by budget officials. They are much more important than "permanent” travel cuts. Influence appears to follow a pattern like that shown in Figure 6.3. Each phase was dominated by groups of budget and department authorities. For example, the DMB budget analysts and the State Budget Director exerted more influence over the allocation of department N reduction targets than the Governor did. Department directors and legislative leaders did not play as active a role as targets are assigned by the State Budget Director. The proportion of influence exerted by the governor on the content of department reduction plans (.2), the influence of department directors on the content of the executive order (.5) and the actual influence of the Governor over the content of the final reduction order (.3) are shown next to the arrows relating each stage of the reduction process to the next in sequence. The individual pie chart for each stage attempts to describe the amount of influence possible by various participants (e.g. the Reduction Order stage is dominated by the Governors recommendations (66 percent) followed by the Quadrant (25 percent) and the appropriations committees (9 percent). m Not all reductions were assigned to departments on a fair share basis (see Appendix D). This research could not attribute this pattern to the influence of any one group of budget officials. Further research is needed to determine if the expected pattern of influence shown in Figure 6.3 is common in those cases where major revenue problems generate permanent program cuts. Such a study should test the hypothesis that the principal budget officer of the Executive Branch exerts more influence over the content of agency budget plans than any other budget official. Ideally, more than one state government should be included in such a study. The hypothesis that rational/comprehensive behavior is a function of the severity and duration of a fiscal crisis also needs to be explored. If sufficient controls can be maintained, such a study might provide answers for those who question the rationality of PPBS budget systems. Figure 6.3 Influence and Reduction Order Content Department Reduction Targets Draft Executive Order A. B. C. D. E. F. G. Department Reduction Plan Reduction Order Division Chiefs / Bureau Chiefs Department Directors DMB Budget Analysts and State Budget Director Governor Legislative Quadrant Legislative Appropriation Committees Other Interest Groups 116 SUMMARY As a result of the experiences of the 1979-1983 budget crisis, the status of programmatic decision making was enhanced. While decisions tended to be made in an incremental way, the process did not exhibit a calculated or pro rata approach to allocation decisions (budget by percentage growth and reduction by fair share). directors became more involved Some department in their agency budget processes. The fiscal crisis seemed to accelerate this process of top management involvement and the role of agency budget officers in program affairs was enhanced in some cases. The framers of the 1960 constitution should not be disappointed with the way that the reduction process worked. They could never have anticipated the degree of fiscal uncertainty evident in the fiscal years between 1979 and 1983, but the process they put in place through Article V worked well. There was reasonable assurance that those programs remaining after 10,000 state workers had been laid off were in fact priority areas. In general those programs cut were the ones that should have been. The transition to a revenue based budgeting process seems to be complete. It appears that government officials are better prepared to respond to fluctuations so common in Michigan's economy. Lessons were painfully learned during 1979-1983 and those officials involved in reduction decisions during that time are not likely to return to the needs based budget policies followed in of the past. might be expected forecasts. in 1986-87, The next down turn according to House Fiscal Agency It should not be necessary to wait long to see if this 117 transition alone will be enough to avoid another crisis like that experienced between 1979 and 1983. APPENDIX A STATE OF MICHIGAN EXECUTIVE ORDER REDUCTION SUMMARY 197 0 -1 7 D e p a rtm e n t o l A g r i c u l tu r e ........................... D e p a rtm e n t o f A tto rn e y G e n e r a l................ D e p a rtm e n t o f Civil R ig h ts ........................... D e p a rtm e n t o l Civil S e r v i c e ......................... D e p a rtm e n t o f C o m m e r c e ............................. D e p a rtm e n t o f C o r r e c t io n s ........................... D e p a rtm e n t o f E d u catio n ............................. E xecutive O f f ic e '................................................ H ig h er E d u catio n C o m m u n ity C o lleg es .................................. S ta le C o lleg es an il U n iv e r s itie s .............. Ju d ic ia rv ................................................................. D e p a rtm e n t o f L abor ...................................... L e g is la tu re ............................................................ L ibrary o f M ic h ig a n ........................................ D e p a rtm e n t o f L icen sin g a n d R e g u la tio n D e p a rtm e n t o f M a n a g e m e n t a n d B u d g et D e p a rtm e n t o f M e n tal H e a lth . . . D e p a rtm e n t o f M ilitary A ffairs . . . D e p a rtm e n t o f N a tu ra l R e so u rces . D e p a rtm e n t o f P ublic H e a lth .. S chool A id .................................... D e p a rtm e n t o f S ocial S ervices . D e p a rtm e n t o f S t a t e .................. D e p a rtm e n t o f S ta te P olice . . . D e p a rtm e n t o f T ra n s p o rta tio n D e p a rtm e n t o f T r e a s u r y ........... D ent S e r v i c e ................................. C a p ital O u tla y & O c h e r . . . . 1 9 7 1 -3 1 9 7 1 -6 1 9 7 4 -1 0 1974-11 1975-11 3 .3 X 1 .0 X 7 0 0 .0 1 0 0 .0 7 0 0 .0 1 0 0 .0 2 ,8 3 8 .7 1 9 8 0 -3 367. 3 2 4 2 .4 1981-7 1961-8 1 9 8 1 -9 9 1 1 -8 1 9 -2 8 -8 1 1 0 -2 2 -8 1 1 ,1 0 0 .0 1 .1 0 0 .0 4 7 5 .0 4 7 5 .6 4 ,0 0 0 .0 5 0 0 .0 1 5 3 .7 2 0 .0 2 0 .0 4 0 0 .0 1 .5 X 5 5 2 .9 9 0 0 .0 9 0 0 .0 2 .9 0 0 .0 9 4 0 .0 4 ,0 0 0 .0 4 .0 X 8 5 5 .6 4 6 .0 1 .2 3 0 .0 5 0 .0 2 .3 0 0 .0 2 0 0 .0 4 .0 Z 3 .5 X 3 .3 X 1 ,3 1 7 .3 1973-12 3 .5 X 4 50. 4 .0 X + 50. 4 .0 X 3 .3 X 3 .3 X 4 .0 X ' 4 0 0 .0 REJECT ED REJECTED REJECTED 7 2 7 .8 4 .0 X 9 4 0 .0 ___1 U J Q J L 5 0 .0 1 ,1 1 3 .4 1 ,1 1 3 .4 6 9 .1 8 9 .1 2 ,0 0 0 .0 5 ,4 9 9 .9 3 ,6 7 2 .3 3 ,8 7 2 .6 6 ,4 1 9 .1 6 ,4 1 9 .1 7 ,4 7 1 .9 7 .5 1 0 .3 6 ,9 0 0 .0 2 5 .7 2 9 .0 1 7 .8 1 7 .3 2 2 .4 1 5 .2 1 .0 X 1 .5 X 6 1 8 .8 5 3 0 .0 5 3 0 .0 1 0 ,0 0 0 .0 1 .5 X 4 .0 X 2 .5 X 15 2 .7 9 6 4 .7 4 .4 3 0 .5 7 0 0 .0 1 .9 0 0 .0 3.5 0 0 .(1 -JJU L U 3 .3 X 3 .0 X 1 .5 0 0 .0 4 , (lull. 0 3 .3 X 4 .0 X 13 1 .4 5 0 .0 5 0 .0 9 ,0 0 0 .1 ) 5 0 0 .0 4 .0 Z 8 3 6 .8 1 ,1 6 3 .6 1 5 ,0 0 0 .0 1 .4 6 2 .4 4 .4 6 0 .0 1 /.6 0 0 .0 1 ,4 6 2 .4 4 .4 0 0 .(1 6 ,5 9 9 .0 5 .5 0 0 .0 3 ,0 0 0 .0 1 5 .1 0 0 .0 • ____ 4 2 7 - P .. 3 ,6 0 0 .0 2 4 ,9 3 7 .0 1 0 ,0 0 0 .0 1 ,3 1 5 .0 1 ,7 0 0 .0 7 2 0 .0 2 0 ,2 4 3 .0 5 2 0 ,0 1 6 ,7 4 3 .0 1 5 0 .0 2 7 0 .0 5 0 .0 2 7 0 .0 3 .3 d 1 .5 X 4 7 ,4 5 0 .0 4 X 3,2 5 0 .o ! 4 0 .4 6 2 .5 1 ,9 0 0 .0 2 0 0 .0 ; X ♦ 1 .6 2 5 . 3 .3 X 3 4 ,0 0 0 .0 7 ,4 0 0 .0 2 1 ,4 6 0 .0 2 1 ,4 0 0 .8 1 5 2 .0 0 0 .0 4 .0 X 1 7 9 .6 1 ,6 6 0 .0 3 .3 X 4 .0 X 8 4 .1 l.O O U .O 1 ,0 0 0 .0 3 ,0 0 0 .0 4 ,5 0 0 .0 1 .0 X 2 .3 X 4 ,0 0 0 .0 2 1 3 .0 2 .8 0 0 .0 1 1 ,9 7 7 .0 2 ,0 0 0 .0 3 2 ,3 0 0 .0 in o .o 9 ,1 0 0 .0 3 2 .2 5 0 .7 7 .7 6 8 .5 5 .3 5 5 .3 340. U 2 4 .0 1 8 .9 4 .2 2 5 .0 1 ,5 0 0 .0 4 .2 2 5 .0 2 7 . f i Jtl.U 2 7 .0 0 0 .0 1 .5 0 0 .0 .... STATE OF MICHIGAN EXECUTIVE ORDER REDUCTION SUMMARY D e p a r tm e n t o l A g r i c u l t u r e .......................... D e p a r tm e n t o f A tto r n e y G e n e r a l .............. D c p a ilm e n l o l C iv il R i g h t s .......................... D e p a r tm e n t o f C iv il S e r v i c e ........................ D e p a r tm e n t o f C o m m e r c e ............................ D e p a r tm e n t o f C o r r e c t i o n s .......................... D e p a r tm e n t o f E d u c a tio n ............................ L 'e c u tiv e O lf ic e , .............. '............................... I lig h e r E d u c a tio n C o m m u n ity C o l l e g e * ................................. .. S la te C o lle g e r a n d U n i v e r s i t i e s .......................... J u d ic ia r y .................................................................... D e p a rt riie n t o f L a b o r ....................................... L e g i s l a t u r e ........................................................... L ie ra rv o f M ic h ig a n .......................................... D e p a r tm e n t o f L ic e n s in g a n d R e g u la tio n D e p a r tm e n t o f M a n a g e m e n t a n d B u d g e t D e p a r tm e n t o f M e n ta l H e a lth ................. D e p a r tm e n t o f M ilita ry A f f a i r s ................... D e p a r tm e n t o f N a tu r a l R e s o u r c e s ............ L T cp artm ent o f P u b lic H e a l t h ..................... S c h o o l A id ........................................................... D e p a r tm e n t o f S o c ia l S e r v i c e s ............................... D e p a r tm e n t o f S t a t e .................................................... D e p a r tm e n t o f S ta le P o l i c e ...................................... D e p a r tm e n t o f T r a n s p o r t a t i o n ............................... D e p a r tm e n t o f T r e a s u r y ............................................. D e n t S e r v i c e .................................................................... C a p ita l O u tl a y & O t h e r .......................................... KO ACTION 1982-9 1982-4 1982-6 1982-13 1981-5 3 -2 5 -8 2 4 -7 -8 2 3 -2 0 -8 2 9 -1 3 -8 2 4 -2 9 -8 3 887.8 . 1 ,6 0 0 .0 1 ,0 0 0 .0 1 ,2 0 0 .0 2 2 0 .0 4 3 3 .4 3 2 3 .0 8 .9 8 8 .1 2 ,8 0 0 .0 7 .7 0 0 .0 1 .4 2 3 .2 1 ,8 0 0 .0 1 .1 2 0 .3 1 3 1 .3 3 0 ,1 8 4 .7 8 0 ,0 0 0 .0 1 .3 8 7 .0 V 1 ,7 0 0 .0 3 0 0 .0 1 3 0 .0 5 0 .0 2 ,4 0 0 .0 4 ,8 0 0 .0 5 ,2 0 0 .0 1 0 0 .0 2 5 0 .0 3 3 .0 2 .3 0 0 .0 . 5 0 .0 1 ,5 0 0 .0 7,00(1.0 2 ,0 0 0 .0 3 .8 0 0 .0 3 2 .0 0 0 .0 27.5O C.0 1 ,3 7 4 .0 1 ,5 0 0 .0 1 .5 0 0 .0 6 ,0 0 0 .0 8 3 1 .0 1 ,1 0 0 .0 2 .8 0 0 .0 3 5 0 .0 1 ,1 0 0 .0 5 0 0 .0 2 .1 0 3 .2 2 4 .0 3 8 .5 3 3 1 .0 2 4 .2 0 0 .0 3 5 0 .0 3 .0 0 0 .0 5 0 .0 2 ,0 2 7 .4 2 ,4 0 0 .0 8 .3 0 0 .0 5 0 .0 2 ,5 0 0 .0 8 .4 0 0 .0 8 3 5 .3 1 2 .1 7 5 .7 4 8 .3 0 0 .0 1 .4 0 0 .0 1 5 ,5 0 0 .0 1 0 .0 0 0 .0 1 0 0 .0 10 0 .0 9 8 0 .4 6 ,8 0 0 .0 4 .2 8 8 .4 3 4 .9 0 0 .0 3 3 ,7 8 8 .7 ----- ------------- 7 2 ,3 0 0 .0 1 7 .9 0 0 ^ 0 "" 2 5 0 .0 4 ,2 0 0 .0 6 .3 0 0 .0 2 5 ,0 0 0 ,0 6 9 ,3 0 0 .0 1 ,2 0 0 .0 5 .0 0 0 .0 -------------- 5 0 .0 3 0 .0 2 5 ,0 0 0 .0 ----- ------------- --- -- - — — ---- - Appendix B Appendix B The reduction process was outlined in the boilerplate or narrative sections of appropriations bills beginning 1975-76. Redaction of expenditures by governor Sec. 10 (1) When it appears to the governor, based upon written information received by him from the director of the department of management and budget and the department of treasury, the actual revenues for a fiscal period will fall below the revenue estimates on which appropriations for that period were based, the estimates being as determined by the legislature in accordance with section 31 of article H of the state constitution of 1963» the governor shall make a finding that actual revenue for that fiscal period, will fall below those revenue estimates. The governor shall then order the director to review all appropriations made by the legislature, except those made for the legislative and judicial branches of government or from funds constitutionally dedicated for specific purposes. (2) Based upon needs, the director of the department of management and budget shall recommend to the governor a reduction of expenditures authorized by the appropriations, either direct or open-ended, for that fiscal year. director The governor shall review the recommendations of the and shall prepare his order 120 containing reductions in 121 expenditures authorized so that actual revenues for the fiscal period will be sufficient to equal the expenditures. not less than 5 d a y s 1 written notice The governor shall give to the members of the appropriations committees of the senate and house specifying a time and place for a joint meeting of the governor and the 2 committees, at which the governor shall present to the committees his recommendations and copies of his proposed order. (3) Not later than 10 days after submission of the order to the committees, each committee by vote of a majority of its members elected and serving shall approve or disapprove the order. Expenditures authorized by appropriations shall not be reduced unless approved by both appropriations committees. Upon approval by both appropriations committees, the director shall carry out and implement the order. (4) If either or both appropriations committees disapproves the order, the order is without force and effect. Not later than 30 days after a proposed order is disapproved, the governor may give reasonable written notice to the time and place of a further joint meeting of the 2 appropriations committees at which time he shall submit another order reducing expenditures authorized by appropriations. Within 10 days after the receipt of that order by the appropriations committees, each appropriations committee shall by a majority of its members elected and serving, approve or disapprove the order. A copy of the order of the governor and resolutions of both the appropriations committees approving it shall be filed with the secretary of state and the order shall become effective. Appendix C MICHIGAN STATE UNIVERSITY S C H O O L O F U R B A N P L A N N I N G A N D L A N D S C A P E A R C H I T E C TURE - EAST LANSING • M I C H I G A N • 48824-1221 [Sample] (Address) Dear (Name): Five years ago, the S tate of Michigan was in the midst of one of the most difficult fiscal crises ever experienced by a sta te government. While most officials are relieved that the worst is over, relatively little is known or documented which might explain how budget reduction decisions were made. A program for dissertation research has been designed to offer some insights into this problem and to suggest how bucket officials decided which programs would be reduced so th at others might be sustained. The purpose of this study is to analyze a select number of decision making situations and to explore an area of budget decision making not studied before. A questionnaire has been prepared to determine how relationships, attitudes, and problems might have effected these decisions. If unexpected differences are evident, explanations and implications for prevailing budget theories will be offered. It is im portant to document what happened in Michigan during the 1979-1983 crisis and to record impressions and interpretations of the way that reduction decisions were reached. With this in mind, I would like you to com plete the enclosed questionnaire and return it to me in the enclosed stamped and addressed envelope. Confidentiality can be assured if you want your responses to remain anonymous. It is hoped th at this research will provide a better understanding and appreciation for what governments go through as budgets are reduced. If you would like, a summary of the results of this stuay can be made available to you or your staff. Thank you very much for your assistance. Sincerely, Kirk L. Lindquist Doctoral Candidate enclosure M S U is an A ffir m a tiv e A c tio n /E q u a l O p p o rtu n ity In stitu tio n Budget Reauction Decisions in State Government 1979-83 Department Budget O fficer Questionnaire: 1. What information proved to be the most useful to you as reduction decisions were made ( between 1979 and 1983 )? 2. Which of the following strategies did your departm ent use to generate required savings? Across the Board Selective (How?) Target and Compromise Cuts made by someone outside the departm ent O ther (specify - _________ 3. Did reduction strategies change as the crisis worsened? If they oio, how and why do you think this happened? 4. Where do you feel th a t most final adjustments to proposed program reduction plans were made? Proposed _______________ Made _______________ Program Level, Agency Top Mgt., DMB, Legislature, or Governor a. Were many changes made? Where did most of these take place? b. At what point in the process were these decisions specifically discussed? 5. Have many of these reductions been restored? If so, who do you feel has been responsible for reversing these earlier decisions? 6. Did the reduction plans change materially from 1979 through to 1981? If so, how 7. Dio some of your departm ent's divisions fair better than others? Was there some characteristic which might explain why these were protected? 8. Do you feel that budget reduction decisions were arrived at in an analytical way? Was there a concern with Program impacts Efficiency Departm ent Missions and Objectives Do you feel this informatipn was used as reduction decisions were reached 9. How would you describe the way DMB prepared and presented Executive Orders to the Legislature? 10. How do you believe issues were selected for the meetings or hearings in which you participated? 11. Do you feel th at funding level decisions are made differently now than they were in the years leading up to the fiscal crisis? How? How have the questions changed? 12. Did your departm ent’s bureaus ever give you an idea of what their program priorities were? If so, was this information useful (was it used) ? How? Comments: Thank-you Budget Reduction Decisions in S tate Government Interview Questions: 1. Do you feel th at groups listed below have much influence over Department program priorities? No influence A Great Deal of Influence 1 2 3 4 5 Client Groups Legislators Agency Directors The Governor 2. 1 1 1 1 ' How much influence do you feel these groups have over Executive policy or budget decisions? No influence A Great Deal of Influence 1 2 3 4 5 Client Groups Legislators Agency D irectors The Governor 3. 1 1 1 1 2 2 2 2 3 3 3 3 4 4 4 4 5 5 5 5 Which group oo you feel is the most influential in budget decisions? Client Groups Agency D irectors 4. 2 3 4 5 2 3 4 5 2 3 4 5 2 3 4 5 __ Legislators __ The Governor What do you feel your role is in budget formulation? Revenue Balancer Budget Reviewer Planner - Prioritizer - R e-prioritizer, Budget Maker Other: ___________ _ 5. Do you agree witn the following perceptions of your responsibilities and the budget process? No Opinion _ ^ f ee Dlsa«r ee Analysis has lim itations but can help in decision making To define government policy Establish expected program performance objectives Analyse alternative means to program performance Select from proposed alternatives Accept but you should not attem pt to shape policy Propose technically sound, accurate and complete solutions to budget problems Influence decisions through good analysis Instruction: 6. For questions 6 and 7, indicate whether you agree (A), disagree (D), or have no opinion (NA). In general, the identified departm ents' Budget Requests contained: (Departments Listed Here) Good Detail Reliable Information Naive Statem ents During the 1980 and 1981 fiscal years, these departm ents were: Legislatively Active Conservative Spenders 7. Which lines of defense were evident in statem ents made by the identified departm ents during the recent budget crisis? a. This program is priceless and should be protected from further cuts b. The effects of proposed reductions in this area cannot be estim ated or measured. c. The effects of proposed cuts will become evident as tim e passes d. "Figures show ..." (good statisics, but not really relevant or useful to you). e. There is no verifyable cause-effect relationship in this area. f. This departm ent will be severely hurt if cuts are made in this area. g. We want you to believe that a relationship exists between what this program does and positive public benefits. „We wish only that we could somehow dem onstrate this to you. Open Endea: 8. What information proved to be the most useful to you as reduction decisions were made ( between 1979 and 1982 )? 9. Did your departm ents use different strategies to generate necessary savings? (P rom pts:____ ___ 10. Across the Board Selective (How?) Target and Compromise Cuts made by someone outside the departm ent Other (specify . . Did reauction strategies change as the crisis worsened? Why do you think this happened? 11. Were programs evaluated for effectiveness during this period (define effectiveness)? If yes, a. b. c. d. 12. How were these topics selected? Were recommendations adopted? Were cost savings realized? Were any programs eliminated as a result of these studies? Where do you feel that most decisions on proposed program funding levels were made? Prompts: Proposed ____________________ Program Level, Agency Top Mgt., Made ____________________ DMB, Legislature, or Governor a. Were many changes made? Where did most of these take place? b. At what point in the process were these decisions specifically discussed? 13. Were many reductions later restored? If so, who do you feel was responsible for getting earlier decisions reversed? 14. Did the reduction plans change m aterially from 1979 through to 1981? How? Prompts: 15. At which point did these changes become apparent? Why do you think that this was the case? Which agencies fairea b e tte r than others? Was there some characteristic which might explain why these were protected? 16. Do you feel that these reduction decisions were arrived at in a particularly analytical or "rational" way? Was tnere a concern with Program impacts Efficiency Department Missions and Objectives Do you feel that this information would have been relevant to these decisions? 17. How woula you describe the way Executive Orders were presented and evaluated by the Legislature? Was a select group instrumental in focussing or filtering these recommended cuts before being considered by the appropriations comm ittees? 18. How do you believe issues were selected for the meetings or hearings in which you participated? 19. As [ ], how concerned were you with the following aspects of the reduction decision making process? Not a t all concerned 1 2 20. 3 4 Very Concerned 5 a. U ltim ate results 12 3 4 5 b. Analysis of Public issues and program Benefits 12 3 4 5 c. The way that targ et reductions were calculated and proposed 12 3 4 5 d. The value and usefulness of budget performance information 12 3 4 5 e. Sequencing, roles and relationships of individuals involved in decision ' making 12 3 4 5 f. To be able to describe the process by which reduction decisions were made 12 3 4 5 g. Explaining how and why reductions were made where they were 12 3 4 5 Do you feel that funding level decisions are made differently now than they were in the years leading up to the fiscal crisis? How? How have the questions changed? 21. Dia the departm ents ever give you an idea of what their program priorities were? If so, was this information useful (was it used) ? How? Demographics: 22. What were your responsibilities in 1980? 23. What are you involved with now (how different from 22)? 24. Were you involved in budget work before January 1978? • Unstructured Section: 25. Would the information gathered through this case stuay be useful for you? 26. If you could ask any question of another person involved in the reduction decisions of the 1980-81, what would it be? Comments: Ideas, or opinions about the subject or this research THANK YOU FOR YOUR COOPERATION Executive Order Suaiary Selected Departments • • !1960-3 Departient n=ll 1981-8 1981-9 1982-4 1982-6 1982-13 1983-5 1 1 1 Average TOTAL E.O. Departient REDUCTION Reduction Average 6F-6P Avg. I Approp. Reduction Stateiide Avg. Z Reduction 1 1 : ! ! 1 i 1 1 i 1 1 6.1 Percent I State Hanaqeient t Budget State Collegas 1 Univ. Public Health Hentai Health Corrections State Police Natural Resources Agriculture Labor Conerce ts:5t:s:a:ss3sx>3>s:::: 1 1 1179.6 (1,000.0 (4,000.0 $1,400.0 1 1981.7 (1,900.0 (4,000.0 (2,600.0 1(6,900.0 (17,817.3 (22,415.2 1(1,382.6 (4,400.0 (5,000.0 (6,300.0 1((,430.5 (200.0 (9,000.0 (24,200.0 1 (727.8 (940.0 (4,000.0 (7,700.0 1 (34.1 (3,000.0 (4,500.0 (15,500.0 1 (886.8 (1,462.4 (5,500.0 (2,400.0 1 (367.J (1,100.0 (4,000.0 (1,600.0 1 (618.8 (530.0 (10,000.0 (1,500.0 1 (552.9 (900.0 (2,900.0 (2,600.0 (100.0 (1,200.0 (1,100.0 (4,100.0 (5,800.0 (32,000.0 (27,500.0 (2,500.0 (6,500.0 (17,900.0 (3,000.0 (5,200.0 (100.0 (100.0 (5,000.0 (50.0 (4,200.0 (1,200.0 (1,700.0 (1,500.0 (6,000.0 (2,400.0 (4,600.0 (7,879.6 (14,684.7 (112,432.5 (26,082.6 (58,730.5 (18,667.8 (28,184.1 (14,499.2 (9,967.3 (20,148.8 (13,952.9 (1,313 (2,447 (18,739 (4,347 (9,788 (3,111 (4,697 (2,417 (1,661 (3,358 (2,325 (10,825 (60,275 (708,425 (93,575 (567,775 (204,050 (126,400 (44,275 (19,900 (39,900 (48,575 12.131 4.06X 2.65Z 4.651 1.89Z 1.52Z 3.721 5.46Z 8.351 B.42Z 4.791 SIj SISVS38SS ai:::sts3itmtfsms3e3ststtastn>8s»33iafs«cz«sis:sssi:«a3S3Csi»«K B»as3»a8s::stsm 33s SSSBS3SS3Ssszssssssxstt M s s in m s s s APPENDIX D SELECTED BIBLIOGRAPHY Anton, Thomas J. The Politics of State Expenditure in Illinois. Urbana, 111., 1966. _________ . "Roles and Symbols in the Determination of State Expend­ itures." Midwest Journal of Political Science. Vol. 11 (February, 1967), 30-31. Appleby, Paul. "The Role of the Budget Division." tration Review. Vol. 17 (Summer, 1957). Public Adminis­ Bailey, John J., and O'Connor, Robert J. "Operationalizing Inerementalism: Measuring the Muddles." Public Administration Review. Vol. 35 (January/February, 1975), 59-61. Banfield, Edward C. "Congress and the Budget: A Planner's Criticism." The American Political Science Review. Vol. M3 (December, 19M9), 1217-1227. Beck, Morris. 1981 . Government Spending; Trends and Issues. New York, N.Y., Bennis, J. Thomas, and Bennis, W., eds. Conflict. Baltimore, MD., 1972. The Management of Change and Berman, Larry. The Office of Management and Budget and the Presidency. Princeton, NJ, 1979. ________ . "0MB and the Hazards of Presidential Staff Work." Administration Review. Vol. 38 (November/December, 1978). Public Biller, Robert P. "Leadership Tactics for Retrenchment." Administration Review. Vol. MO (November/December, 1980). Public Borcherding, Thomas. Budgets and Bureaucrats:__The Sources o_f-Govern­ ment Growth. Durham, N.C., 1977. Boskin, Michael J., and Wildavsky, Aaron, eds. The Federal Government: Economics and Politics. San Francisco, CA., 1982. Braybrooke, David, and Lindblom, Charles E. Glencoe, 111., 1963. 131 A Strategy of Decision. 132 Brazer, Harvey. "Anatomy of a Fiscal Crisis: The Michigan Case." Public Budgeting and Finance. Vol. 2 (Winter, 1980), 130-142. ________ , and Laren, Deborah S., eds. Structure. Ann Arbor, MI 1982. Michigan's Fiscal and Economic Breit, William. "Starving the Leviathan: Balanced Budget Prescriptions Before Keynes." Fiscal Responsibility in Constitutional Democracy. James M. Buchanan and Richard E. Wagner, eds. Boston, MA, 1978. Brown, David S. "The Staff Man Looks in the Mirror." Public Administration Review. Vol. 23 (June, 1963), 67-73. Buck, Arthur E. The Budget in Governments of Today. New York, NY, 1934. ________ . Public Budgeting. New York, NY, 1929. Burkhead, Jesse. Government Budgeting. New York, NY, 1956. Caiden, Naomi. "Patterns of Budgeting." Vol. 38 (November/December, 1978). Public Administration Review. Cleveland, Frederick A. "Evolution of the Budget Idea in the United States." Annals of the American Academy of Political and Social Science. Vol. 62 (November, 1915), 15-35. _______ , and Breck, Arthur E. New York, NY, 1920. The Budget and Responsible Government. Committee on Government Operations, Subcommittee Hearings, U.S. House of Representatives. Improving Federal Budgeting and Appropriations. 85th congress, 1st Session. Washington, D.C., 1957. Committee, Joint Economic, Subcommittee Hearings, The PlanningProgramming-Budgeting System:__ P.cogr&aa- antLJteftsntlalg. Congress, 1st Session. Washington, D.C., 1967. 90th Congressional Budget Office. Reducing the Federal Budget: Strategies and Examples:__Fiscal Years 1982-1986. Washington, D.C., 1981. ________ . Reducing the Federal Budget: ■■Stratejgies..ariA-_0.P_tlCin3. Washington, D.C., 1982. Cope, Glen, and Grubb, W. Norton. "Restraint in a Land of Plenty, Revenue and Expenditure Limitations in Texas." Public Budgeting and Finance. Vol. 2 (Winter, 1982), 143-156. Crecine, John P. "The Role of 0MB in Executive Branch Budgetary Decision Making." in Crecine, John P., ed., Research in Policy, and Management. Greenwich, Ct., 1981. 133 _________ • Government Problem Solving; A Computer Simulation of Municipal-.Budgeting. Chicago, 111., 1969. Cyert, Richard and March, James. Englewood Cliffs, N.J., 1963. Dahl, Robert and Lindblom, Charles. New York, N.Y., 1953. A Behavioral Theory of the Firm. Politics. Economics, and Welfare. Davis, OHO A., M.A.H. Dempster, and Aaron Wildavsky, BA Theory of the Budgetary Process," The American Political Science Quarterly. Vol. 60 (September, 1966) pp. 529-547. Dempster, M.A.H. and Wildavsky, Aaron. Magic Size for An Increment." (September, 1979), 371-389. "On Change: Or There is No Political Studies. Vol. 27 Downs, Anthony. "Why the Government Budget is Too Small in a Democ­ racy. World Politics. Vol. 12 (July 1960), 541-563. _________ . Inside Bureaucracy. Boston, MA, 1967. Dunconibe, Sydney and Kinney, Richard. "Cutbacks - Idaho Style." Budgeting and Finance. Vol. 4 (Summer 1984), 87-98. Fenno, Richard F., Jr. Congressmen in Committees. Public Boston, MA, 1973- _________ . "The House Appropriations Committee as a Political System: The Problem of Integration." The American Political Science Review. Vol. 56 (June 1962), 310-324. . The Power of the Purse. Boston, MA, 1966. Fisher, Frank. Politics. Values and Public Policy. Boulder, Co., 1980. Fisher, Louis. "Congress, the Executive, and the Budget." Annals of the American Academy of Political and Social Sciences. Vol. 411 (1974), 102-113. _________ . Presidential Spending Power. Princeton, N.J., 1975. Fitzpatrick, Edward A. 1918 . Budget Making in a Democracy. New York, N.Y., Gerwin, Donald. "Towards a Theory of Budget Decision Making." Adminis­ trative Science Quarterly. Vol. 14 (March 1969), 41. Gist, John. "Increment and Base in the Congressional Appropriations Process." American Journal of Political Science. Vol. 21 (May 1977), 341-352. 13k Friedrick, Carl, ed. 196k. Nomos VII: Rational Decision. New York, N.Y., Golembiewski, Robert T. and Rubin, Jack, eds. Public Budgeting and Finance: Readings in Theory and Practice. Hasca, 111., 1975. Greider, William. nThe Education of David Stockman." Monthly (December 1981), 27-5k. The Atlantic Guhde, Robert and Mustapha, Husain. "Budget Making in Ohio: A Test of the Process Model." Western Political Quarterly. Vol. 3k (December 1981), 578-592. Hammond, Thomas H., and Knott, Jack H. A Zero-Based Look at Zero Based Budgeting; Or Why Its Failures in State Government are Being Duplicated in Washington. New Brunswick, N.J., 1979. Harper, Edwin L., Kramer, Fred A., and Rouse, Andrew M. "Implementation and Use of PPB in Sixteen Federal Agencies." Public Administration Review. Vol. 29 (November/December 1969). Haveman, Joel. Congress and the Budget. Bloomington, IN, 1978. Haveman, Robert H. and Margolis, Julius, eds. Policy Analysis. Chicago, 111. 1977. Public Expenditure and Havens, Harry S. "MBO and Program Evaluation, Or What Ever Happened to PPBS?" Public Administration Review. Vol. 36 (January/February 1976). Hoover Commission on the Organization of the Executive Branch of the Government. Budgeting, and Accounting. Washington, D.C. 19k9House Fiscal Agency, House of Representatives. State of Michigan, Review of.the State of Michigan Fl3cal-Budgetarv Process. Lansing, MI, 1976. Hrebrenar, Ronald J. "the Utility Role of Budget Decision Making." Western Political Quarterly. Vol. 29 (December 1976), 575-588. Kamlet, M.S., and Mowery, D.C. "The Budgetary Base in Federal Resource Allocation." American Journal of Political Science. Vol. 2k (1980), 80k-821. Key, V.O. "The Lack of Budgetary Theory." Review. Vol. 3k (19k0), 1138. American Political Science Korb, Lawrence. "The Budgetary Process in the Department of Defense 19k7-1977: The Strengths and Weaknesses of Three Systems." Public Administration Review. Vol. 37 (July/August 1977), 33k-336. 135 Lanth, Thomas P. "Zero Based Budgeting in Georgia State Government: Myth and Reality." Public Administration Review. Vol. 38 (September/October 1978). Levine, Charles H., Rubin, Irene, and Walohojian, George. "Managing Organization Retrenchment." Administration and Society. Vol. 14 (May 1982), 105. _________ . "More on Cutback Management: Hard Questions for Hard Times." Public Administration Review. Vol. 39 (March/April 1979), 179-183. _________ . "Organizational Decline and Cutback Management." Public Administration Review. Vol. 38 (July/August 1978), 316-325. Lewis, Carol and Logalbo, Anthony. "Cutback Principles and Practices: A Checklist for Managers." Public Administration Review. Vol. 40 (March/April 1980). Lewis, Vern. "Toward a Theory of Budgeting." Review. Vol. 12 (Winter 1952). Public Administration LeLoup, Lance T. "The Myth of Incrementalism: Analytical Choices in Budgetary Theory." Policy Vol. 10 pp. 488-509. ______. Budgetary Politics. Brunswick, Ohio, 1977. _________ . "Explaining Agency Appropriations Change, Success, and Legislative Support: a Comparative Study of Agency Budget Determination." Unpublished Ph.D. Dissertation. Columbus, Ohio, 1973. Lijphart, Arend, "Comparative Politics and Comparative Method," American Political Science Review. Vol. 65, (1971) PP. 687-693. Lindblom, Charles E. N.J., 1968. The Policy Making Process. _________ . "The Science of Muddling Through." Review. Vol. 19 (Spring 1959), 79-88. March, James G. and Simon, Herbert A. 1958. Miller, David W. and Starr Martin K. Englewood Cliffs, N.J., 1967. Englewood Cliffs, Public Administration Organizations. New York, N.Y., The Structure of Human Decisions. Mosher, Frederick C. Program Budgeting; Theory and Practice, with a Particular Reference to the U.S. Department of the. Army. Chicago, 111. 1954. 136 Nakamura, Robert T. and Smallwood, Frank. tion. New York, N.Y., 1980. The Politics of Implementa­ Natchez, Peter B., and Irvin C. Bupp, "Policy and Priority in the Budgetary Process," The American Political. Science Review. Vol. 67 (1973), PP. 951-963. Neustadt, Richard E. "Presidency and Legislation: The Growth of Central Clearance." The American Political Science Review. Vol. 48 (September 1954), 641-671. Nienaber, Jeanne and Wildavsky, Aaron. The Budgeting and Evaluation of Federal Recreation Programs or Money Doesn't Grow on Trees. New York, N.Y., 116-142. Niskanim, William A. Bureaucracy and Representative Government. Chicago, 111. 1971. ________ . "Bureaucrat and Politicians." Vol. 18 (December 1975), 617-644. Journal of Law and Economics. _________ . Structural Reform of the Federal Budget Process. Washington, D.C. 1973. Novick, David. CA, 1976. Origin and History of Program Budgeting. Santa Monica, Oates, Wallace E., ed. The Political Economy of Fiscal Federalism. Lexington, MA, 1977. Pearson, Norman M. "The Budget Bureau: From Routine Business to General Staff." Public Administration_Review. Vol. 3 (Spring 1943). Pierce, Lawrence. The Politics of Fiscal Policy Formulation. Englewood Cliffs, N.J., 1971. Pressman, Jeffrey and Wildavsky, Aaron. 1983. Implementation. Burkley, CA, Pyhrr, Peter A. Zero Base. Budgeting; A Practical Management Tool for Evaluation Expenses. New York, N.Y., 1973* Rose, Richard. "Implementation and Evaporation: The Record of MBO." Public Administration Re-Vlew. Vol. 37 (January/February 1977). Schick, Allen, ed. Perspectives on Budgeting. Washington, D.C., 1980. __________. "How the Budget was Won and Lost" in Orustein, Norman J., ed., President and Congress.: Assessing Regan's First Year. Washington, D.C., 1981. 137 _________ . "Budgetary Adaptations in Resource Scarcity." in Levine, C.H. and Rubin, I., eds., Fiscal Stress and Public Policy. Beverly Hills, CA., 1980. _________ . Budget Innovation in the States. Washington, D.C., 1980. ________ . "The Budget Bureau that was: Thoughts on the Rise, Decline and Future of a Presidential Agency." Law and Contemporary Problems. Vol. 35 (Summer 1970), 519-539. ________ . "Systems Politics and Systems Budgeting." Public Administration Review. Vol. 29 (March/April 1969), 42. _________ . "The Road to PPB: Stages of Budgetary Reform." Administration Review. Vol. 26 (December 1966). Public Schulman, Paul. "Nonincremental Policy Making: Notes Toward an Alter­ native Paradigm." The American Political Science Review. Vol. 49 (December 1975),.1355. Schultz, Charles L. The. Politics and Economics of PublicSpending. Washington, D.C., 1968. Seligman, Lester. "Presidential Leadership: The Inner Circle and Institutionalization." Journal of Politics. Vol. 18 (August 1956), 410-426. Sharkansky, Ira and Turnball, Augustino M. "Budget Making in Georgia and Wisconsin: A Test of a Model." Midwest Journal of Political Science. Vol. 13 (November 1969). Sharkansky, Ira, "Agency Requests, Gubernatorial Support and Budgetary Success in State Legislatures," The American Political Science Review. Vol. 62 (1968), pp. 1220-1231. Shepsle, Kenneth A. "The Failure of Congressional Budgeting." The Center for the Study of America Business. Washington University Publication 46 (March, 1982). Simon, Herbert A. N.Y., 1957. Simon Herbert A. _________ . 1960. Administrative Behavior. 2nd Edition. New York, The Sciences of the Artificial. Cambridge, MA, 1969. The New Science of Management Decision. _________ . "Special Issue on Decision Making." Quarterly. Vol. 3 (December 1958). New York, N.Y., Administrative Science Smithies, Arthur. The Budgetary Process in the United States. York, N.Y., 1955. New 138 Strausman, Jeffrey D., and Bozeman, Barry. "Shrinking Budgets and the Shrinkage of Budget Theory." Public Administration Review. Vol, 42 (November/December 1982), 509-515. Strausman, Jeffrey D. "Spending More and Enjoying it Less." Comparative Politics (January 1981), 235-251. Sullivan, Gerald E. "Incremental Budget Making in the American States: A Test of the Anton Model." Journal of Politics. Vol. 34 (May 1972), 643. Symposium. "Government Decision Making." Vol. 24 (September 1964), 153-165. _________ . "Budget Theory." (Winter 1950), 20-31. Public Administration Review. Public Administration Review. Vol. 10 _________ . "Performance Budgeting: Has the Theory Worked?" Public Administration Review. Vol. 20 (Spring 1960), 63-85. Waldo, Dwight. The Administrative State. New York, N.Y., 1948. Walker, Robert. "William A. Jump: The Staff Officer as a Personality." Public Administration Review. Vol. 14 (Autumn 1954), 233-246. Weinberg, Martha Wagner. Wildavsky, Aaron. _________ . Managing the State. Cambridge, MA, 1977. Budgeting. Boston, MA, 1975. How To, Limit Public Spending. Los Angeles, Ca., 1980. _________ . "A Budget For All Seasons? Why the Traditional Eudget Lasts." Public Administration Review. Vol. 38 (November/December 1978), 501-509. -------- . "Rescuing Policy Analysis from PPBS." Public Administration Review. Vol. 29 (March/April 1969), 189-202. . "The Political Economy of Efficiency: Cost-Benefit Analysis, Systems Analysis and Program Budgeting." Administration Review. Vol. 26 (December 1966), 292-310. Public ________ . "The Political Implications of Budgetary Reform." Administration-Revlew. Vol. 21 (Autumn 1961), 183-190. Public ________ . "If Planning is Everything, Maybe Sciences. Vol. 4 (1973), 127-153. Policy Its Nothing." _________ . "The Self Evaluating Organization." Public Administration Review. Vol. 32 (September/October 1972), 509-520. 139 _________, and Hamond, Arthur. "Comprehensive Versus Incremental Budgeting In the Department of Agriculture." Administrative Science Quarterly. Vol. 10 (December 1965), 321-346. --------- . Politics of the Budgetary Process. Boston, MA, 1964. Willouby, William F. The National Budget System. Baltimore, MD, 1927. --------- • The_.Movement for Budgetary Reform in the States. New York, NY, 1918.